Search

CN-122024511-A - Accurate-positioning reverse car-searching navigation method for parking lot and electronic equipment

CN122024511ACN 122024511 ACN122024511 ACN 122024511ACN-122024511-A

Abstract

The method comprises the steps of extracting landmark objects from structured map data, and calculating the saliency score of each landmark object according to the spatial distribution density and the heat value of the landmark objects; the method comprises the steps of responding to a parking space recording request of a user, obtaining a space position coordinate of the parking space where the user is located, searching at least one target landmark object with a significance score larger than a preset scoring threshold value in a preset searching radius, establishing and storing a space association relation between the parking space and the target landmark object to obtain a navigation path association relation, and selecting landmark objects of different types from the landmark objects of the navigation path association relation according to the distance between the user and the space position coordinate of the parking space to generate a navigation instruction. By implementing the method, the vehicle searching efficiency of the reverse vehicle searching of the parking lot of the user can be improved.

Inventors

  • WU SONGTAI
  • YANG XIAOTONG

Assignees

  • 北京车易付科技股份有限公司

Dates

Publication Date
20260512
Application Date
20260209

Claims (10)

  1. 1. The reverse car searching navigation method for the precisely positioned parking lot is characterized by being applied to electronic equipment, and comprises the following steps: Obtaining structured map data of a parking lot, wherein the structured map data comprises parking space positions, channel layout, elevator positions and safety exit positions; extracting landmark objects from the structured map data, and acquiring user activity data of peripheral areas of each landmark object in a preset time period to obtain a heat value of each landmark object; calculating a significance score of each landmark object according to the spatial distribution density of the landmark objects and the heat value; Responding to a parking space recording request of a user, acquiring a space position coordinate of the parking space where the user is located, retrieving at least one target landmark object with the significance score larger than a preset scoring threshold value in a preset searching radius, establishing and storing a space association relation between the parking space and the target landmark object, and obtaining a navigation path association relation; and selecting landmark objects of different types from landmark objects in the association relation of the navigation paths according to the distance between the user and the space position coordinates of the parking space to generate a navigation instruction.
  2. 2. The method according to claim 1, wherein the step of calculating a saliency score for each of the landmark objects based on the spatial distribution density and the heat value of the landmark objects comprises: Counting the number and average distance of other landmark objects in a preset detection radius by taking the landmark object as a center, and calculating the spatial distribution density value of the landmark object according to the number and the average distance; Acquiring type attributes of the landmark objects, determining corresponding weight coefficients according to the type attributes, and multiplying the heat value by the weight coefficients to obtain weighted heat values; and carrying out weighted summation on the spatial distribution density value and the weighted heat value to obtain a significance score of each landmark object.
  3. 3. The method according to claim 1, wherein the step of retrieving at least one target landmark object with the significance score greater than a preset scoring threshold within a preset search radius, establishing and storing a spatial association relationship between the parking space and the target landmark object, and obtaining a navigation path association relationship specifically includes: searching the landmark objects in the preset searching radius by taking the space position coordinates of the parking space as the center, and screening landmark objects with the significance scores larger than a preset scoring threshold value from the searched landmark objects to obtain a candidate landmark object set; The candidate landmark object sets are arranged in a descending order according to the saliency scores, grouping is carried out according to type attributes, and each group of landmark objects with the arrangement being larger than the preset number of preset ranks are selected to serve as the target landmark objects; Calculating azimuth angle and linear distance between the space position coordinates of the parking spaces and each target landmark object, judging the visibility of the target landmark object relative to the space position coordinates of the parking spaces, and marking the visibility judgment result as visible direct vision or invisible direct vision; and carrying out association storage on the space position coordinates of the parking spaces, the target landmark object, the azimuth angle, the linear distance and the visibility judgment result to obtain the association relation of the navigation paths.
  4. 4. The method according to claim 3, wherein the step of selecting different types of landmark objects from the landmark objects of the navigation path association relation according to the distance between the user and the space position coordinates of the parking space to generate the navigation instruction specifically comprises: acquiring the current position coordinates of the user, and calculating the real-time distance between the current position coordinates and the space position coordinates of the parking space; Determining a landmark granularity level according to the real-time distance, determining the landmark granularity level as a regional level when the real-time distance is greater than a first preset threshold, determining the landmark granularity level as a channel level when the real-time distance is not greater than the first preset threshold and is greater than a second preset threshold, and determining the landmark granularity level as a local level when the real-time distance is not greater than the second preset threshold; Extracting landmark objects with type attributes matched with the landmark granularity level from the navigation path association relation, screening out directly visible landmark objects according to the visibility judgment result when the extracted landmark objects comprise a plurality of landmark objects, and selecting the landmark object with the highest significance score if the directly visible landmark objects do not exist; calculating the relative azimuth between the current position coordinate of the user and the selected landmark object, and judging the steering direction by combining the azimuth angle; And generating a navigation instruction according to the landmark object, the relative azimuth, the steering direction and the real-time distance, wherein the navigation instruction comprises a landmark name, a direction guide and a distance prompt.
  5. 5. The method of claim 4, wherein after the step of selecting a different type of landmark object from the landmark objects of the navigation path association according to the distance between the user and the space position coordinates to generate a navigation instruction, the method further comprises: Acquiring real-time position coordinates of the user, and calculating a real-time azimuth angle between the real-time position coordinates and the space position coordinates of the parking space; Comparing the real-time azimuth angle with azimuth angles stored in the association relation of the navigation path, and calculating an angle deviation value between the real-time azimuth angle and the azimuth angles; when the angle deviation value is larger than a preset deviation threshold value, judging that the user deviates from a preset navigation path, and searching a landmark object in a preset correction radius by taking the real-time position coordinate as the center; Screening the landmark object with the highest saliency score and the direct vision as a correction guiding landmark object from the retrieved landmark objects according to the visibility judging result; and calculating a corrected azimuth angle and a corrected distance based on the relative position relation between the corrected guiding landmark object and the space position coordinates of the parking space, generating a corrected navigation instruction and outputting the corrected navigation instruction to the user.
  6. 6. The method according to claim 1, wherein after the step of selecting a different type of landmark object from among landmark objects of the navigation path association according to the distance between the user and the space position coordinates of the parking space to generate a navigation instruction, the method further comprises: Acquiring position stay data of the user in the navigation process, wherein the position stay data comprises stay position coordinates and stay time length; Judging whether the stay position coordinates are located in a preset recognition range of the target landmark object, and marking the corresponding target landmark object as a landmark object with low recognition degree when the stay position coordinates are located in the preset recognition range and the stay time exceeds a preset time threshold; Counting the accumulated times of marked low-recognition landmark objects in a preset counting period, and performing weight reduction processing on the heat value of the low-recognition landmark objects according to a preset weight reduction coefficient when the accumulated times exceed a preset time threshold; And calculating the update significance score of the low-recognition landmark object based on the reduced heat value, removing the low-recognition landmark object from the navigation path association relation when the update significance score is lower than the preset score threshold value, and searching for the alternative landmark object within the preset search radius to supplement the navigation path association relation.
  7. 7. The method according to claim 1, wherein after the step of selecting a different type of landmark object from among landmark objects of the navigation path association according to the distance between the user and the space position coordinates of the parking space to generate a navigation instruction, the method further comprises: generating a plurality of candidate description modes of the same target landmark object, wherein the candidate description modes comprise color feature description, number identification description and space position description; respectively applying different candidate description modes to navigation instructions of different users, and recording the time spent by each user from receiving the navigation instruction to reaching the space position coordinates of the parking space; Counting the average time consumption corresponding to each candidate description mode, and selecting the shortest candidate description mode in the average time as the optimal description mode; And storing the optimal description mode in association with the target landmark object, and describing the target landmark object by adopting the optimal description mode when a navigation instruction containing the target landmark object is generated.
  8. 8. An electronic device comprising one or more processors and memory coupled to the one or more processors, the memory to store computer program code comprising computer instructions that the one or more processors invoke the computer instructions to cause the electronic device to perform the method of any of claims 1-7.
  9. 9. A computer readable storage medium comprising instructions which, when run on an electronic device, cause the electronic device to perform the method of any of claims 1-7.
  10. 10. A computer program product, characterized in that the computer program product, when run on an electronic device, causes the electronic device to perform the method of any of claims 1-7.

Description

Accurate-positioning reverse car-searching navigation method for parking lot and electronic equipment Technical Field The application relates to the field of electric digital data processing, in particular to a precisely-positioned reverse car searching navigation method for a parking lot and electronic equipment. Background With the acceleration of the urban process and the continuous increase of the conservation quantity of motor vehicles, the scale of underground parking lots of large commercial complexes, transportation hubs and public facilities is continuously enlarged, and the number of parking spaces is thousands or even tens of thousands. Parking lot internal structure is getting more and more complex, including multiple floors, multiple partitions and intricate channel networks. After parking, users often face the trouble that the position of a parking space is difficult to memorize and the direction of the vehicle is lost when returning to the vehicle searching, so that the vehicle searching time is overlong, and the travel efficiency and the user experience are affected. Therefore, providing efficient and accurate reverse car-seeking navigation services for parking lots to users is an urgent need. In the prior art, a reverse car-searching navigation system of a parking lot records the parking position coordinates of a user by locating equipment and establishing an electronic map in the parking lot, and plans a walking route from the current position of the user to a parking space based on a shortest path algorithm when the user initiates a car-searching request. The system presents the planned route to the user in the form of arrow indication or text description, and guides the user to go to the target parking space along the calculated shortest path. This navigation approach relies on abstract coordinate information and geometric path calculations to provide distance and direction based instructions to the user such as "turn right 50 meters forward" or "walk along the a-channel to the B-region". However, in the actual environment of a large parking lot, the user lacks visual knowledge of the abstract coordinate position and spatial orientation, and it is difficult to correlate the distance value and direction description in the navigation instruction with the surrounding actual environment. The interior environment of the parking lot is relatively single, and lacks obvious visual differences, and a large number of similar columns, signs and channels cause confusion to users. When the user receives the instruction of 50 meters forwards, the distance is difficult to accurately judge under the condition of lacking an obvious reference object, and when the user is informed of right turning to enter the C area, if a plurality of similar area inlets exist around, the user cannot quickly identify the correct turning position. The navigation mode based on the abstract coordinates and the geometric path causes frequent pauses, repeated confirmation and even wrong route passing of the user in the process of executing the navigation instruction, has low vehicle searching efficiency and poor user experience. Disclosure of Invention The application provides a precisely-positioned reverse car searching navigation method for a parking lot and electronic equipment, which are used for improving the car searching efficiency of reverse car searching of a parking lot for a user. In a first aspect, the application provides a precisely positioned reverse car-searching navigation method for a parking lot, which is applied to electronic equipment, and comprises the steps of obtaining structured map data of the parking lot, wherein the structured map data comprises a parking space position, a channel layout, an elevator position and a safety exit position; the method comprises the steps of obtaining structured map data, extracting landmark objects from the structured map data, obtaining user activity data of surrounding areas of each landmark object in a preset time period to obtain a heat value of each landmark object, calculating saliency scores of each landmark object according to space distribution density and heat values of the landmark objects, responding to a parking space record request of a user, obtaining space position coordinates of a parking space where the user is located, retrieving at least one target landmark object with the saliency scores being larger than a preset scoring threshold value in a preset searching radius, establishing space association relation between the parking space and the target landmark object, storing the space association relation to obtain a navigation path association relation, and selecting landmark objects of different types from the landmark objects of the navigation path association relation according to the distance between the user and the space position coordinates to generate navigation instructions. In the above embodiment, the electronic device extracts landmark obje