Parking point selection and operation method and device based on positioning scene recognition

Through positioning scene recognition and heat map optimization, the positioning accuracy problem of fixed-point parking in scenarios with poor signal quality is solved, efficient and accurate parking in different environments is achieved, and user experience is improved.

CN119028169APending Publication Date: 2024-11-26XIAOAN KEJI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411126005.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In the scenario of poor signal quality, the existing fixed-point parking technology has reduced positioning accuracy, resulting in poor user experience.

Method used

Through the method of positioning scene recognition, the vehicle positioning data of the target area is determined, the positioning scene recognition is carried out, and the positioning heat map is generated. Parking points are screened based on the heat map and vehicle flow data is optimized to ensure that the parking point layout and operation strategies are ensured in different environments Achieve efficient and accurate fixed-point parking.

Benefits of technology

In an environment with poor signal quality, through intelligent identification and differentiated response strategies, the accuracy and user experience of designated parking are improved, and convenient facilities are ensured not to disturb the public.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119028169A_ABST
    Figure CN119028169A_ABST
Patent Text Reader

Abstract

The invention provides a parking spot selection and operation method and device based on positioning scene recognition. The method comprises the following steps: determining vehicle positioning data of a target area; performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; based on the positioning performance of the target area corresponding to the scene recognition result, parking point selection and / or parking point operation are / is performed on the target area, and the defects that in a traditional electric bicycle operation scheme, a fixed-point parking method is not suitable for scenes where signals are prone to being interfered and the signal quality is poor, the positioning precision is reduced, and the user experience is poor are overcome. According to the technical scheme of the invention, the defects of poor user experience and the like in the prior art are overcome, efficient and accurate fixed-point parking of the user in various environments can be realized through differential coping strategies in different scenes, and the vehicle use experience of the user is optimized on the premise of ensuring that convenience facilities do not disturb residents.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a parking point selection and operation method and device based on positioning scene recognition. Background Art

[0002] Shared electric motorcycles, as a representative of green travel, have played an important role in solving urban traffic problems. However, with the surge in the number of vehicles, the problem of disorderly parking of shared electric motorcycles has become increasingly prominent, which not only affects the daily travel experience, but also has a negative impact on urban management and the appearance of the city. In response to this problem, fixed-point parking technology has been introduced to regulate user parking behavior and improve the orderliness of parking.

[0003] However, current fixed-point parking technology mainly relies on vehicle positioning systems and electronic fences at parking spots, which guide users to park their vehicles in preset parking frames to achieve orderly management. However, in actual applications, the method of drawing electronic fences based solely on offline parking frame positions has obvious limitations. In the case of poor signal quality, the positioning accuracy is greatly reduced, thus affecting the accuracy of fixed-point parking and user experience. Summary of the invention

[0004] The present invention provides a parking point selection and operation method and device based on positioning scene recognition, which is used to solve the problem that the fixed-point parking method in the prior art is not suitable for scenes where the signal is easily interfered with and the signal quality is poor, which will lead to a decrease in positioning accuracy and a poor user experience. By controlling the parking points for specific scenes, the user experience can be improved.

[0005] The present invention provides a parking point selection and operation method based on positioning scene recognition, comprising: Determine vehicle location data in the target area; Performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; Based on the positioning performance of the target area corresponding to the scene recognition result, parking point selection and / or parking point operation are performed on the target area.

[0006] According to a parking point selection and operation method based on positioning scene recognition provided by the present invention, the scene recognition result includes the area type of the target area, and the positioning performance of the target area corresponding to the scene recognition result is used to select a parking point and / or operate the parking point for the target area, including: Determine a positioning heat map of the target area based on the area type of the target area, the area location and hardware facilities of the target area; the positioning heat map is used to reflect the positioning performance of each preset parking point in the target area; Based on the positioning heat map, parking point selection and / or parking point operation are performed in the target area.

[0007] According to a parking point selection and operation method based on positioning scene recognition provided by the present invention, the parking point selection for the target area based on the positioning heat map includes: Determine the traffic flow data of each preset parking spot in the target area; Based on the positioning heat map and the traffic flow data, a target parking spot is screened out from various preset parking spots in the target area for the user to park.

[0008] According to a parking point selection and operation method based on positioning scene recognition provided by the present invention, the parking point operation is performed on the target area based on the positioning heat map, including: Acquire monitoring images of each preset parking spot in the target area; Based on the monitoring image and the positioning heat map, parking point operations are performed in the target area.

[0009] According to a parking point selection and operation method based on positioning scene recognition provided by the present invention, the positioning scene recognition is performed based on the vehicle positioning data to obtain a scene recognition result, including: Inputting the vehicle positioning data into a positioning scene recognition model to obtain a scene probability distribution output by the positioning scene recognition model; Determining a scene recognition result based on the scene probability distribution; The positioning scene recognition model is trained based on sample vehicle positioning data and area type labels of the sample vehicle positioning data.

[0010] According to a parking point selection and operation method based on positioning scene recognition provided by the present invention, the parking point operation includes adjusting the area of ​​the return buffer zone corresponding to each preset parking point in the target area, and / or increasing or decreasing the return methods corresponding to each preset parking point.

[0011] The present invention also provides a parking point selection and operation device based on positioning scene recognition, comprising: A data determination unit, used to determine vehicle positioning data in a target area; A scene recognition unit, used to perform positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; A parking point control unit is used to select and / or operate parking points in the target area based on the positioning performance of the target area corresponding to the scene recognition result.

[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a parking point selection and operation method based on positioning scene recognition as described in any one of the above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described parking point selection and operation methods based on positioning scene recognition.

[0014] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-described parking point selection and operation methods based on positioning scene recognition.

[0015] The parking point selection and operation method and device based on positioning scene recognition provided by the present invention perform positioning scene recognition according to the vehicle positioning data of the target area to obtain a scene recognition result, and perform parking point selection and / or parking point operation on the target area based on the positioning performance of the target area corresponding to the scene recognition result, thereby realizing intelligent scene recognition of specific areas and optimization of parking point layout and operation strategy, overcoming the defect that the fixed-point parking method in the traditional electric motorcycle operation scheme is not suitable for scenarios where the signal is easily interfered and the signal quality is poor, resulting in reduced positioning accuracy and poor user experience. Through differentiated response strategies under different scenarios, users can achieve efficient and accurate fixed-point parking in various environments, and optimize the user experience of the car while ensuring that the convenience facilities do not disturb the public. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 It is a flow chart of a parking point selection and operation method based on positioning scene recognition provided by the present invention; Figure 2 is a schematic diagram of the parking point selection process provided by the present invention; Figure 3 It is a schematic diagram of the parking point operation process provided by the present invention; Figure 4 is a schematic diagram of a positioning scene recognition process provided by the present invention; Figure 5It is a structural schematic diagram of a parking point selection and operation device based on positioning scene recognition provided by the present invention; Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In order to solve the problem of disorderly parking of shared electric motorcycles caused by the surge in the number of shared electric motorcycles, major car companies have introduced fixed-point parking technology, aiming to regulate user parking behavior through this technology and improve the orderliness of shared electric motorcycle parking.

[0020] However, the current fixed-point parking technology mainly relies on the vehicle positioning system and the electronic fence of the parking spot, and achieves orderly management by guiding users to park their vehicles in the preset parking frame. However, in practical applications, the method of drawing electronic fences based solely on the location of offline parking frames has obvious limitations. Specifically, due to the complex and changeable urban environment, the environments in which different parking frames are located are significantly different, which directly affects the signal quality and positioning accuracy of the positioning device. In an open environment without obvious obstruction, the positioning device can work stably and provide high-precision location information; however, in obstructed environments such as "urban canyons" (special environments formed by dense high-rise buildings and narrow streets), the positioning signal is easily interfered, which can easily lead to a significant decrease in positioning accuracy, thereby affecting the accuracy of fixed-point parking and user experience.

[0021] In this regard, the present invention provides a parking point selection and operation method based on positioning scene recognition, which aims to perform positioning scene recognition on the target area, and optimize the parking point layout and operation strategy according to the positioning performance corresponding to the scene recognition result, to ensure efficient and accurate fixed-point parking in various environments, and to optimize the user's car experience while ensuring that the convenience facilities do not disturb the public. Figure 1 is a flow chart of a parking point selection and operation method based on positioning scene recognition provided by the present invention, such as Figure 1 As shown, the method includes: Step 110, determining vehicle positioning data in the target area; Step 120, performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; Step 130 , based on the positioning performance of the target area corresponding to the scene recognition result, parking spot selection and / or parking spot operation are performed on the target area.

[0022] Specifically, considering the current operation of electric motorcycles, when the fixed-point parking technology is introduced to control the parking of vehicles, the impact of the environment in which the parking frame is located on the fixed-point parking effect is not taken into account, resulting in large positioning deviations and inconvenience in returning the vehicle in scenarios with poor signal quality, resulting in poor user experience. In an embodiment of the present invention, intelligent recognition and differentiated response strategies for different scenarios are proposed. When drawing the parking frame, the environment in which the parking frame is located is re-sensed, and parking point management is performed based on this, so that users can return the vehicle conveniently even in scenarios with poor signal quality, poor positioning environment, and large offset.

[0023] It is understandable that in the actual operation process, before the parking point control is carried out, the first thing that needs to be determined is the area that needs to be controlled, that is, the area to be controlled in the motorcycle operation area, which is called the target area. The target area here can be one or more. The determination of the target area can be based on the background data of the motorcycle operation platform, for example, the user feedback that the signal quality in a certain area is poor, the motorcycles are parked in a messy way, etc., or it can be determined based on the information of the staff's field investigation, and each partition in the operation area can be directly used as the target area, that is, the parking point control can be carried out on all areas in the operation area on a regular basis (every half a month, a month, etc.), and the embodiment of the present invention does not make specific limitations on this.

[0024] After determining the target area, the embodiment of the present invention can perform the positioning scene recognition task to perceive the environment of the parking frame, so as to better manage the parking spot. Before that, it is necessary to obtain the positioning data in the target area, that is, the vehicle positioning data in the target vehicle, so as to perform scene perception based on the vehicle positioning data, so as to obtain the specific scene of the target area, such as tree-lined avenues, uncovered main roads, "urban canyons" between high-rise buildings, etc.

[0025] Among them, vehicle positioning data refers to the positioning data of the vehicle in the target area within a historical period of time (such as half a month, one month, three months, half a year, etc.). Specifically, it can be the positioning data uploaded when the vehicle passes through the target area and needs to report location information. It can also be the positioning data fed back by the vehicle to the platform when the user parks in the target area. It can also be positioning data related to vehicle positioning obtained by other means, for example, positioning data of the vehicle in the target area obtained through a public network platform when positioning. The embodiments of the present invention do not make specific limitations on this.

[0026] Specifically, after obtaining the vehicle positioning data, in an embodiment of the present invention, the vehicle positioning data can be used to perform positioning scene recognition to identify the specific scene in which the target area is located, thereby obtaining a scene recognition result for the target area. The scene recognition result here may only include the type of the specific scene in which the target area is located, that is, the area type of the target area, such as a boulevard, an unobstructed main road, an "urban canyon" between high-rise buildings, etc., or may include both the area type and the confidence level, that is, the various area types and their probabilities to which the target area may correspond. In this case, scene recognition actually obtains a probability distribution, which includes various area types and their corresponding confidence levels. The area type in which it is located can be determined based on the confidence level.

[0027] Furthermore, after obtaining the scene recognition result of the target area, in an embodiment of the present invention, parking point management and control can be performed based on this scene recognition result to optimize the parking point selection, layout, and operation strategy of the parking point, so that users can return the vehicle conveniently when riding in the target area, and ensure accuracy and efficiency when parking at a fixed point, thereby optimizing the user's car experience.

[0028] Specifically, here we can first determine the positioning performance of the target area based on the scene recognition result of the target area, that is, the positioning performance of the target area can be determined according to the area type in the scene recognition result obtained by the positioning scene recognition. The positioning performance here refers to the quality of positioning when the vehicle is positioned in the target area, which includes the strength of the positioning signal, the size of the positioning deviation, etc.

[0029] Among them, the positioning performance of the target area can be specifically determined by a pre-set relationship list between the area type and the positioning performance, that is, the area type that matches the area type of the target area can be found from this relationship list, and the positioning performance of the target area can be directly determined based on the positioning performance corresponding to this area type. Further, in an embodiment of the present invention, after determining the positioning performance corresponding to this area type, it is also possible to determine the positioning performance of the target area based on the positioning performance corresponding to this area type, combined with relevant information of the vehicle positioning data of the target area, such as the signal strength of the positioning data reported by the vehicle, the transmission time of the reported positioning data, etc., or it can be based on the positioning performance corresponding to this area type, combined with the regional location of the target area, environmental information, etc., such as signal towers, base stations, and large-volume infrastructure in the area, to determine the positioning performance of the target area.

[0030] After determining the positioning performance of the target area, in the embodiment of the present invention, the parking point control of the target area can be carried out according to the positioning performance. That is, the parking point layout optimization and / or operation control of the target area can be carried out according to the various indicators of the positioning of the vehicle in the area reflected by the positioning performance of the target area, such as the accuracy, the degree of offset, the difficulty, the positioning time, etc. Here, it can be specifically selected for the target area, for example, a certain area or one / more small areas are selected from the target area as parking points, the positions of the parking points pre-set in the target area are adjusted, parking points with strong positioning performance are selected from the parking points pre-set in the target area, and parking points with weak positioning performance are screened out from the parking points pre-set in the target area, etc., or parking point operation can be carried out for the target area, for example, the operation mode, operation means, etc. of the parking points pre-set in the target area are increased, decreased, changed, etc., and the above two can also be performed at the same time, that is, parking point selection and parking point operation are carried out for the target area.

[0031] The parking point selection and operation method based on positioning scene recognition provided by the present invention performs positioning scene recognition according to the vehicle positioning data of the target area to obtain a scene recognition result, and performs parking point selection and / or parking point operation on the target area based on the positioning performance of the target area corresponding to the scene recognition result, thereby realizing intelligent scene recognition of specific areas and optimization of parking point layout and operation strategy, and overcoming the defect that the fixed-point parking method in the traditional electric motorcycle operation scheme is not suitable for scenarios where the signal is easily interfered and the signal quality is poor, resulting in reduced positioning accuracy and poor user experience. Through differentiated response strategies under different scenarios, users can achieve efficient and accurate fixed-point parking in various environments, and optimize the user experience of the car while ensuring that the convenience facilities do not disturb the public.

[0032] Based on the above embodiment, the scene recognition result includes the area type of the target area; Step 130 includes: Based on the area type of the target area, as well as the area location and hardware facilities of the target area, a positioning heat map of the target area is determined; the positioning heat map is used to reflect the positioning performance of each preset parking point in the target area; Based on the positioning heat map, parking spot selection and / or parking spot operation are performed in the target area.

[0033] Specifically, in step 130, the process of selecting a parking spot and / or operating a parking spot for the target area according to the positioning performance of the target area corresponding to the scene recognition result specifically includes: First, the positioning heat map of the target area can be determined based on the scene recognition results obtained by the positioning scene recognition for the target area. That is, based on the area type in the scene recognition results, combined with the location, environment and other information of the target area, a comprehensive measure is made to determine the positioning heat map of the target area. The positioning heat map here is a heat map that can reflect the positioning performance of various locations in the target area, which can be determined based on the area type of the target area, the area location of the target area, and the conditions of the hardware facilities in the target area (location, type, etc.).

[0034] It should be noted that before parking point control is performed in the target area, parking points can be set in the target area. At this time, the parking points can be understood as preliminarily determined parking points, that is, preset parking points. In the subsequent control process, the preset parking points are also targeted to facilitate parking point selection and / or parking point operation. Based on this, it can be determined that the above positioning heat map actually reflects the positioning performance of each preset parking point in the target area.

[0035] Specifically, the positioning heat map of the target area can be determined according to the area type of the target area, the location of each preset parking point in the target area, and the hardware facilities near each preset parking point. Here, specifically, according to the area type of the target area, the area type that matches it can be determined from a preset list of relationships between area types and positioning performance, and then, based on the positioning performance corresponding to this area type, the location of each preset parking point in the target area and the hardware facilities near each preset parking point, for example, whether there are base stations, signal towers, large coverings, etc. near each preset parking point, are combined to determine the positioning heat map of the target area.

[0036] Subsequently, parking spot selection and / or parking spot operation can be performed for the target area based on this positioning heat map. Specifically, the parking spot layout optimization and / or operation control can be performed for the target area based on various indicators of the positioning of vehicles in the area at various preset parking spots as reflected by the positioning heat map of the target area, such as accuracy, offset, difficulty, positioning time, etc.

[0037] Here, specifically, the target area may be selected based on the positioning performance of each preset parking point in the positioning heat map. For example, the layout of each preset parking point in the target area may be adjusted based on the positioning performance of each preset parking point. Specifically, the position of the preset parking point may be adjusted, or some preset parking points with weak positioning performance may be screened out, or some preset parking points with strong positioning performance may be selected. The embodiment of the present invention does not specifically limit this. The target area may also be operated based on the positioning performance of each preset parking point in the positioning heat map. For example, the area of ​​the return buffer zone in the preset parking point may be adjusted based on the positioning performance of each preset parking point, a new return method (such as Bluetooth lock return, radio frequency identification return, etc.) may be added, etc. The target area may also be selected and operated. The embodiment of the present invention does not specifically limit this.

[0038] Based on the above embodiment, the parking point selection in the target area is performed based on the positioning heat map, including: Determine the traffic flow data of each preset parking spot in the target area; Based on the positioning heat map and traffic flow data, the target parking spot is selected from the preset parking spots in the target area for users to park.

[0039] Specifically, the process of selecting parking spots in the target area according to the positioning heat map may include: Taking into account the operation of electric motorcycles, the setting of parking spots is fundamentally to ensure that the vehicles are parked stably and orderly to facilitate users to return the vehicles. The convenience of returning the vehicles not only needs to consider the positioning performance, but also the location of the parking spot, that is, the traffic volume nearby. If there is no parking spot near the area with heavy traffic, then the user will need to walk a long way or take a detour when returning the vehicle, which is very inconvenient and will lead to a poor user experience.

[0040] In view of this, in the embodiment of the present invention, the traffic flow data of the target area is taken into consideration when selecting parking spots in parking spot management and control, so as to improve the convenience of returning the car while ensuring good positioning performance, thereby optimizing the user experience. That is, the traffic flow data of each preset parking spot in the target area can be determined first. The traffic flow data here can be obtained by analyzing the parking data and vehicle usage data of each preset parking spot in the area in the operation platform in the recent period of time.

[0041] After determining the traffic flow data, in an embodiment of the present invention, the traffic flow data and the positioning heat map can be combined to select parking points for the target area. That is, the target parking point can be screened out from the preset parking points in the target area according to the positioning heat map and the traffic flow data as the final selected parking point for the user to park.

[0042] Figure 2 is a schematic diagram of the parking point selection process provided by the present invention, such as Figure 2 As shown, the target parking point is selected from the preset parking points based on various indicators of the positioning heat map when the vehicle is positioned at each preset parking point, such as accuracy, offset, difficulty, positioning time, etc., and the traffic volume and peak period near each preset parking point in the target area reflected by the traffic volume data. Specifically, a preset parking point with strong positioning performance, high ranking of traffic volume among the preset parking points, and relatively stable traffic volume can be selected from the preset parking points as the target parking point, so that the user can return the car after use to ensure the convenience of returning the car.

[0043] Based on the above embodiment, parking point operation is performed in the target area based on the positioning heat map, including: Acquire monitoring images of each preset parking spot in the target area; Based on monitoring images and positioning heat maps, parking point operations are carried out in the target area.

[0044] The parking point operation includes adjusting the area of ​​the return buffer zone corresponding to each preset parking point in the target area, and / or increasing or decreasing the return modes corresponding to each preset parking point.

[0045] Specifically, the process of operating parking spots in the target area according to the positioning heat map may include: First, the monitoring images of each preset parking spot in the target area can be obtained. The monitoring images here are images obtained by real-time monitoring of each preset parking spot through the monitoring cameras near each preset parking spot. Various information of each preset parking spot can be reflected through this monitoring image, such as the area information of the return buffer zone (such as the area size, the location of the parking frame, etc.), vehicle information (such as whether the vehicle is parked in a standardized and orderly manner), user return information (such as return time, duration, peak period, etc.), etc.

[0046] Next, based on this monitoring image and the positioning heat map of the target area, parking point operations can be performed in the target area. Specifically, the monitoring results of the target area can be determined based on the monitoring image. The monitoring results here can include any one or more of the identification results of whether the parking of vehicles in the target area is orderly, the detection results of whether the return of the vehicle is convenient, and the analysis results of whether the return buffer zone can accommodate vehicles returned during peak hours. Then, based on this monitoring result and in combination with the positioning heat map of the target area, parking point operations can be performed in the target area. The parking point operations here can include the adjustment of the area of ​​the return buffer zone corresponding to each preset parking point in the target area, and / or the increase or decrease of the return methods corresponding to each preset parking point in the target area.

[0047] Figure 3 It is a schematic diagram of the parking point operation process provided by the present invention, such as Figure 3 As shown, the area of ​​the return buffer corresponding to each preset parking point in the target area can be adjusted and / or a new return method can be added based on the monitoring results determined by the monitoring image and the positioning heat map. For example, when it is confirmed that the return buffer corresponding to a preset parking point cannot accommodate all vehicles returned at the parking point during the peak period, the area of ​​the corresponding return buffer can be appropriately and timely increased based on the positioning heat map. For another example, when it is confirmed that it is difficult to return a vehicle in the return buffer corresponding to a preset parking point, a new return method can be added, such as Bluetooth lock return, radio frequency identification return, etc., to facilitate users to return the vehicle.

[0048] In addition to this, in an embodiment of the present invention, after obtaining the monitoring results and positioning heat maps, artificial intelligence technology, such as a multimodal large model, can also be used to perform parking point operations on preset parking points in the target area, so as to obtain rectification suggestions given by the model for the preset parking points in the target area.

[0049] Based on the above embodiment, step 120 includes: Inputting the vehicle positioning data into the positioning scene recognition model to obtain the scene probability distribution output by the positioning scene recognition model; Based on the scene probability distribution, determine the scene recognition result; The positioning scene recognition model is trained based on sample vehicle positioning data and area type labels of the sample vehicle positioning data.

[0050] Specifically, in step 120, the process of performing positioning scene recognition based on the vehicle positioning data to obtain the scene recognition result can be specifically implemented through machine learning, that is, in the absence of a true value device, based on a machine learning method, the vehicle positioning data can be used to identify the specific scene of the target area to obtain the scene recognition result of the target area.

[0051] Figure 4 is a schematic diagram of the positioning scene recognition process provided by the present invention, such as Figure 4 As shown, a model of a machine learning scenario, i.e., an initial model, can be pre-built, and then the initial model can be trained to obtain a trained model for parking scene recognition, i.e., a positioning scene recognition model, by using pre-collected sample vehicle positioning data and the area type labels of the sample vehicle positioning data. Here, the collection of samples during model training, the determination of labels, and the calculation of losses during the training process can all be determined by existing methods.

[0052] It should be noted that, during training, the collected sample vehicle positioning data can also be subjected to data processing operations such as screening, cleaning, and classification, and then the processed data can be used for model training.

[0053] After the positioning scene recognition model is obtained through offline training, in an embodiment of the present invention, online recognition can be performed based on this positioning scene recognition model, that is, by deploying an entity (such as a cloud platform, server, etc.) with a trained positioning scene recognition model, and using the vehicle positioning data of the target area, positioning scene recognition is performed to obtain the scene probability distribution output by the model.

[0054] The scene probability distribution here includes the various area types that the target area identified by the model may correspond to (such as tree-lined avenues, uncovered main roads, "urban canyons" between high-rise buildings, etc.), as well as the probability corresponding to each area type.

[0055] After the scene probability distribution is obtained, in the embodiment of the present invention, the scene recognition result of the target area can be determined based on the scene probability distribution. That is, the final scene recognition result of the target area is determined according to the probability that the target area corresponds to each area type as reflected in the scene probability distribution. That is, the area type with the highest probability is selected from each area type as the final area type of the target area, and the scene recognition result of the target area can be determined based on this area type, or this area type and the corresponding probability.

[0056] The parking point selection and operation device based on positioning scene recognition provided by the present invention is described below. The parking point selection and operation device based on positioning scene recognition described below and the parking point selection and operation method based on positioning scene recognition described above can be referenced to each other.

[0057] Figure 5 is a structural schematic diagram of a parking point selection and operation device based on positioning scene recognition provided by the present invention, such as Figure 5 As shown, the device comprises: A data determination unit 510, used to determine vehicle positioning data in a target area; A scene recognition unit 520 is used to perform positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; The parking point control unit 530 is used to select parking points and / or operate parking points in the target area based on the positioning performance of the target area corresponding to the scene recognition result.

[0058] The parking point selection and operation device based on positioning scene recognition provided by the present invention performs positioning scene recognition according to the vehicle positioning data of the target area to obtain a scene recognition result, and performs parking point selection and / or parking point operation for the target area based on the positioning performance of the target area corresponding to the scene recognition result, thereby realizing intelligent scene recognition of specific areas and optimization of parking point layout and operation strategy, and overcoming the defect in traditional electric motorcycle operation schemes that the fixed-point parking method is not suitable for scenarios where signals are easily interfered with and the signal quality is poor, resulting in reduced positioning accuracy and poor user experience. Through differentiated response strategies under different scenarios, users can achieve efficient and accurate fixed-point parking in various environments, and optimize the user experience of the vehicle while ensuring that convenient facilities do not disturb the public.

[0059] Based on the above embodiment, the scene recognition result includes the area type of the target area, and the parking point control unit 530 is used to: Determine a positioning heat map of the target area based on the area type of the target area, the area location and hardware facilities of the target area; the positioning heat map is used to reflect the positioning performance of each preset parking point in the target area; Based on the positioning heat map, parking point selection and / or parking point operation are performed in the target area.

[0060] Based on the above embodiment, the parking point control unit 530 is used to: Determine the traffic flow data of each preset parking spot in the target area; Based on the positioning heat map and the traffic flow data, a target parking spot is screened out from various preset parking spots in the target area for the user to park.

[0061] Based on the above embodiment, the parking point control unit 530 is used to: Acquire monitoring images of each preset parking spot in the target area; Based on the monitoring image and the positioning heat map, parking point operations are performed in the target area.

[0062] Based on the above embodiment, the scene recognition unit 520 is used for: Inputting the vehicle positioning data into a positioning scene recognition model to obtain a scene probability distribution output by the positioning scene recognition model; Determining a scene recognition result based on the scene probability distribution; The positioning scene recognition model is trained based on sample vehicle positioning data and area type labels of the sample vehicle positioning data.

[0063] Based on the above embodiment, the parking point operation includes adjusting the area of ​​the vehicle return buffer zone corresponding to each preset parking point in the target area, and / or increasing or decreasing the vehicle return methods corresponding to each preset parking point.

[0064] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute a parking point selection and operation method based on positioning scene recognition, the method comprising: determining vehicle positioning data of a target area; performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; and performing parking point selection and / or parking point operation on the target area based on the positioning performance of the target area corresponding to the scene recognition result.

[0065] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0066] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the parking point selection and operation method based on positioning scene recognition provided by the above-mentioned methods. The method includes: determining vehicle positioning data of a target area; performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; and performing parking point selection and / or parking point operation for the target area based on the positioning performance of the target area corresponding to the scene recognition result.

[0067] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the parking point selection and operation method based on positioning scene recognition provided by the above-mentioned methods. The method includes: determining vehicle positioning data of a target area; performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; and performing parking point selection and / or parking point operation for the target area based on the positioning performance of the target area corresponding to the scene recognition result.

[0068] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0069] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A parking point selection and operation method based on positioning scene recognition, characterized in that: include: Determine vehicle location data in the target area; Performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; Based on the positioning performance of the target area corresponding to the scene recognition result, parking point selection and / or parking point operation are performed on the target area.

2. The parking point selection and operation method based on positioning scene recognition according to claim 1 is characterized in that: The scene recognition result includes the area type of the target area, and the selecting and / or operating a parking spot for the target area based on the positioning performance of the target area corresponding to the scene recognition result includes: Determine a positioning heat map of the target area based on the area type of the target area, the area location and hardware facilities of the target area; the positioning heat map is used to reflect the positioning performance of each preset parking point in the target area; Based on the positioning heat map, parking point selection and / or parking point operation are performed in the target area.

3. The parking point selection and operation method based on positioning scene recognition according to claim 2 is characterized in that: The selecting of parking spots in the target area based on the positioning heat map includes: Determine the traffic flow data of each preset parking spot in the target area; Based on the positioning heat map and the traffic flow data, a target parking spot is screened out from various preset parking spots in the target area for the user to park.

4. The parking point selection and operation method based on positioning scene recognition according to claim 2 is characterized in that: The performing parking point operation on the target area based on the positioning heat map includes: Acquire monitoring images of each preset parking spot in the target area; Based on the monitoring image and the positioning heat map, parking point operations are performed in the target area.

5. The parking point selection and operation method based on positioning scene recognition according to any one of claims 1 to 4, characterized in that: The performing positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result includes: Inputting the vehicle positioning data into a positioning scene recognition model to obtain a scene probability distribution output by the positioning scene recognition model; Determining a scene recognition result based on the scene probability distribution; The positioning scene recognition model is trained based on sample vehicle positioning data and area type labels of the sample vehicle positioning data.

6. The parking point selection and operation method based on positioning scene recognition according to any one of claims 1 to 4, characterized in that: The parking point operation includes adjusting the area of ​​the vehicle return buffer zone corresponding to each preset parking point in the target area, and / or increasing or decreasing the vehicle return modes corresponding to each preset parking point.

7. A parking point selection and operation device based on positioning scene recognition, characterized in that: include: A data determination unit, used to determine vehicle positioning data in a target area; A scene recognition unit, used to perform positioning scene recognition based on the vehicle positioning data to obtain a scene recognition result; A parking point control unit is used to select and / or operate parking points in the target area based on the positioning performance of the target area corresponding to the scene recognition result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the parking point selection and operation method based on positioning scene recognition as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the parking point selection and operation method based on positioning scene recognition as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the parking point selection and operation method based on positioning scene recognition as described in any one of claims 1 to 6 is implemented.