Automatic cruise parking method and system based on marked precision map

Abstract

The application discloses an automatic cruise parking method and system based on a marked precision map, and the method comprises the following steps: collecting lane environment images in a parking lot by an image collection device arranged on a vehicle during vehicle driving; acquiring position information of lane elements in the parking lot according to the lane environment images, wherein the lane elements comprise a lane, an intersection and a parking space; constructing a topological relation graph between each lane and the intersection according to the position information of the lane and the intersection; mounting the parking space to the corresponding lane according to the position information of the parking space according to the topological relation graph, and generating a marked precision map; and automatically cruising based on the marked precision map, and parking the vehicle in the parking lot. The application is favorable for saving the cost of automatic cruising and parking, and is favorable for large-scale laying of an automatic cruise parking system.

Description

Technical Field

[0001] This invention relates to the field of automatic cruise parking, and more particularly to an automatic cruise parking method and system based on a high-resolution map. Background Technology

[0002] As people's demand for driving technology increases, automated cruise parking technology has gradually developed. Currently, automated cruise parking systems mainly operate in two directions: one involves the parking lot providing map information and the vehicle's location within the parking lot, along with route planning; the other involves the vehicle building and storing high-precision map information of the parking lot, with the vehicle performing map positioning during the automated cruise parking process. The problems with existing technologies are that the parking lot-provided map method requires certain requirements for the parking lot, necessitating the establishment of dedicated cruise routes and an additional vehicle positioning system, resulting in high maintenance costs for the parking lot and hindering large-scale deployment. The vehicle-based method, which involves building and maintaining a high-precision parking lot map, requires detailed lane, obstacle, and parking space information. When a vehicle enters the map area, centimeter-level precision positioning is required, and route planning is based on the high-precision map, placing high demands on the accuracy of vehicle positioning. Therefore, existing technologies rely on either the parking lot-based or vehicle-based high-precision map positioning system, which is not conducive to saving costs for automated cruise parking and hinders large-scale deployment of automated cruise parking systems.

[0003] Therefore, existing technologies still need improvement and development. Summary of the Invention

[0004] The main objective of this invention is to provide an automatic cruise parking method and system based on a high-precision map, which aims to solve the problem that existing technologies rely on high-precision map positioning systems at the parking lot or vehicle end, which is not conducive to saving the cost of automatic cruise parking and to the large-scale deployment of automatic cruise parking systems.

[0005] To achieve the aforementioned objective, a first aspect of the present invention provides an automatic cruise parking method based on a high-precision map, wherein the automatic cruise parking method based on the high-precision map includes:

[0006] During vehicle operation, images of the lane environment within the parking lot are captured by image acquisition devices installed on the vehicle.

[0007] The location information of lane elements in the parking lot is obtained based on the lane environment image. The lane elements include: lanes, intersections, and parking spaces.

[0008] Based on the location information of the lanes and the intersections, a topological relationship diagram between each lane and the intersection is constructed;

[0009] Based on the topology diagram, the parking spaces are attached to the corresponding lanes according to their location information to generate a high-resolution map.

[0010] The vehicle automatically cruises based on the high-precision map and parks itself in the parking lot.

[0011] Optionally, constructing a topological relationship diagram between each lane and the intersection based on the location information of the lanes and the intersections includes:

[0012] As the vehicle enters the parking lot entrance, it continuously collects lane environment images within the parking lot through image acquisition equipment, performs image recognition on the lane environment images, and obtains the currently perceived lane lines.

[0013] The vehicle travels straight along the current perceived lane line. When the image recognizes an intersection, the perceived lane line between the parking lot entrance and the intersection is taken as the current lane.

[0014] The vehicle travels straight along the next sensing lane line. When the image recognizes the next intersection, the next sensing lane line between the next intersection and the previous intersection is taken as the next lane, and the vehicle is controlled to continue driving and collect lane environment images until the parking lot exit is recognized.

[0015] Obtain the location information of the lanes and intersections between the parking lot entrance and the parking lot exit, and construct a topological relationship diagram between each lane and the intersection.

[0016] Optionally, the step of attaching the parking space to the corresponding lane based on the location information of the parking space according to the topology graph to generate a high-resolution map includes:

[0017] During vehicle operation, image recognition is performed on the real-time collected lane environment images to obtain parking space lines;

[0018] Based on the location information of the parking space line, determine whether the parking space corresponding to the parking space line is set in the current lane where the vehicle is traveling;

[0019] When the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, the parking space is attached to the corresponding lane in the topology graph according to the location information of the parking space, and a high-precision map is generated.

[0020] Optionally, determining whether the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, based on the position information of the parking space line, includes:

[0021] If the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, and it is determined that the lane where the parking space is located is available for parking, then according to the location information of the parking space, the parking lot information and the topology diagram, the parking space is attached to the lane corresponding to the topology diagram;

[0022] If the parking space corresponding to the parking space line is not located in the current lane where the vehicle is traveling, it is determined that the lane where the parking space is located is not suitable for parking, and the lane will be discarded during this loading process.

[0023] Optionally, after generating a high-resolution map by attaching the parking space to the corresponding lane based on the location information of the parking space according to the topology graph, the method further includes:

[0024] A preset high-precision map, wherein the preset high-precision map is a high-precision map corresponding to a known lane environment;

[0025] The vehicle's image acquisition device acquires lane environment images, and the lane environment images are compared with preset scene descriptors to determine whether the vehicle is driving on a lane in a preset high-precision map. The scene descriptor is the feature information of the lane environment in the preset high-precision map obtained through neural network learning.

[0026] If the vehicle is traveling in a lane in the preset standard definition map, then determine whether to update the parking spaces in the preset standard definition map based on the lane environment image;

[0027] If the vehicle is not traveling in a lane of the preset precision map, a new precision map is generated based on the lane environment image.

[0028] Optionally, the automatic cruise based on the high-precision map, whereby the vehicle parks within the parking lot, includes:

[0029] The vehicle cruises based on the high-precision map;

[0030] Based on the parking spaces on the high-precision map, the system uses sensors to determine in real time whether the parking spaces are vacant.

[0031] If the parking space is vacant, then the parking space will be used as the target parking space for parking.

[0032] If the parking space is not vacant, the vehicle will cruise based on other parking spaces on the high-resolution map and their location information until it identifies a target parking space and parks there.

[0033] A second aspect of the present invention provides an automatic cruise parking system based on a high-precision map, the system comprising:

[0034] The image acquisition module is used to acquire images of the lane environment in the parking lot through image acquisition devices installed in the vehicle during vehicle operation;

[0035] The lane element acquisition module is used to acquire the location information of lane elements in the parking lot based on the lane environment image. The lane elements include: lanes, intersections, and parking spaces.

[0036] The topology graph construction module is used to construct a topology graph between each lane and the intersection based on the location information of the lanes and the intersections;

[0037] The high-precision map generation module is used to attach the parking space to the corresponding lane according to the topology diagram and the location information of the parking space to generate a high-precision map;

[0038] An automatic cruise parking module is used for automatic cruise based on the high-precision map, in which the vehicle parks itself in the parking lot.

[0039] Optionally, the topology graph construction module includes:

[0040] The current perceived lane line acquisition unit is used when the vehicle enters the parking lot entrance, continuously acquires lane environment images in the parking lot through image acquisition equipment, performs image recognition on the lane environment images, and obtains the current perceived lane line.

[0041] The current lane acquisition unit is used for the vehicle to travel straight along the current perceived lane line. When the image recognizes an intersection, the perceived lane line between the parking lot entrance and the intersection is taken as the current lane.

[0042] The next lane acquisition unit is used for the vehicle to travel straight along the next sensing lane line. When the image recognizes the next intersection, the next sensing lane line between the next intersection and the previous intersection is taken as the next lane, and the vehicle is controlled to continue to drive and collect lane environment images until the parking lot exit is recognized.

[0043] A relationship graph construction unit is used to obtain the location information of the lanes and intersections between the parking lot entrance and the parking lot exit, and to construct a topological relationship graph between each lane and the intersection.

[0044] A third aspect of the present invention provides a smart terminal, the smart terminal including a memory, a processor, and an automatic cruise parking program based on a high-precision map stored in the memory and executable on the processor, wherein when the automatic cruise parking program based on the high-precision map is executed by the processor, it implements the steps of any one of the automatic cruise parking methods based on a high-precision map.

[0045] As can be seen from the above, in this invention, during vehicle operation, an image acquisition device installed on the vehicle acquires lane environment images within the parking lot; based on the lane environment images, the location information of lane elements within the parking lot is obtained, the lane elements including: lanes, intersections, and parking spaces; based on the location information of the lanes and intersections, a topological relationship diagram between each lane and the intersection is constructed; based on the topological relationship diagram, the parking space is attached to the corresponding lane according to the location information of the parking space, generating a high-precision map; based on the high-precision map, the vehicle automatically cruises and parks within the parking lot.

[0046] Compared with existing technologies, this invention provides an automated cruise parking method based on a high-precision map. The method obtains the location information of lane elements within the parking lot from lane environment images. These lane elements include lanes, intersections, and parking spaces. A topological relationship graph is constructed between the lanes and intersections. Based on the topological relationship graph, the location information of the parking spaces is mapped to the corresponding lanes to generate a high-precision map. The vehicle then automatically parks within the parking lot based on this high-precision map. This method of generating a high-precision map based on topological relationships does not rely on high-precision map positioning systems at the parking lot or vehicle end, which helps save on the cost of automated cruise parking and facilitates the large-scale deployment of automated cruise parking systems. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart illustrating an automatic cruise parking method based on a high-precision map provided in an embodiment of the present invention.

[0049] Figure 2 This is an embodiment of the present invention. Figure 1 A detailed flowchart of step S300 is shown below;

[0050] Figure 3 This is a schematic diagram of a process for constructing a topology graph according to an embodiment of the present invention;

[0051] Figure 4 This is an embodiment of the present invention. Figure 1 A detailed flowchart of step S400 is shown below;

[0052] Figure 5 This is a schematic diagram of a process for mounting a parking space according to an embodiment of the present invention;

[0053] Figure 6 This is a schematic diagram of a cruise process for mounting a parking space provided by an embodiment of the present invention;

[0054] Figure 7 This is a high-precision map provided in an embodiment of the present invention;

[0055] Figure 8 This is a schematic diagram of the automatic cruise parking system based on a high-precision map provided in an embodiment of the present invention;

[0056] Figure 9 This is an embodiment of the present invention. Figure 8 A schematic diagram of the specific structure of the topology graph construction module 630;

[0057] Figure 10 This is a block diagram illustrating the internal structure of a smart terminal provided in an embodiment of the present invention. Detailed Implementation

[0058] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0059] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0060] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0061] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0062] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to classification." Similarly, the phrases "if determined" or "if classified to [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once classified to [the described condition or event]," or "in response to classification to [the described condition or event]."

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0065] As people's demand for vehicle driving increases, automatic cruise parking technology is gradually developing. In the existing technology, automatic cruise parking systems mainly have two directions: one is that the parking lot provides map information and the vehicle's position in the parking lot and provides route planning for the car; the other is that the vehicle establishes and stores high-precision map information of the parking lot, and the vehicle performs map positioning during the automatic cruise parking process.

[0066] The existing technologies have several problems. Firstly, maps provided by the parking lot require specific parking facilities, including dedicated cruise routes and additional vehicle positioning systems, resulting in high maintenance costs and hindering large-scale deployment. Secondly, vehicles need to build and maintain high-precision parking lot maps, which require detailed lane, obstacle, and parking space information. When a vehicle enters the map area, it needs centimeter-level positioning accuracy and route planning based on the high-precision map, placing high demands on vehicle positioning accuracy. Therefore, existing technologies rely on either parking lot or vehicle-side high-precision map positioning systems, which is detrimental to cost savings in automated cruise parking and hinders large-scale deployment of automated cruise parking systems.

[0067] To address at least one of the aforementioned problems, this invention involves the following steps: During vehicle operation, image acquisition devices installed on the vehicle capture lane environment images within a parking lot; based on the lane environment images, location information of lane elements within the parking lot is obtained, including lanes, intersections, and parking spaces; based on the location information of the lanes and intersections, a topological relationship diagram between each lane and intersection is constructed; based on the topological relationship diagram and the location information of the parking spaces, the parking spaces are attached to the corresponding lanes, generating a high-resolution map; and based on the high-resolution map, the vehicle automatically cruises and parks within the parking lot.

[0068] Compared with existing technologies, this invention provides an automated cruise parking method based on a high-precision map. The method obtains the location information of lane elements within the parking lot from lane environment images. These lane elements include lanes, intersections, and parking spaces. A topological relationship graph is constructed between the lanes and intersections. Based on the topological relationship graph, the location information of the parking spaces is mapped to the corresponding lanes to generate a high-precision map. The vehicle then automatically parks within the parking lot based on this high-precision map. This method of generating a high-precision map based on topological relationships does not rely on high-precision map positioning systems at the parking lot or vehicle end, which helps save on the cost of automated cruise parking and facilitates the large-scale deployment of automated cruise parking systems.

[0069] Exemplary methods

[0070] like Figure 1 As shown, this embodiment of the invention provides an automatic navigation method based on a high-precision map. Specifically, the method includes the following steps:

[0071] Step S100: During vehicle operation, lane environment images within the parking lot are captured by an image acquisition device installed on the vehicle.

[0072] In this embodiment, before constructing the target map, the vehicle needs to drive along a road, and image acquisition devices installed on the vehicle are used to capture lane environment images within the parking lot. Furthermore, in one application scenario, the lane environment images are uploaded to the cloud and combined with neural networks for image recognition and analysis.

[0073] Step S200: Obtain the location information of lane elements in the parking lot based on the lane environment image. The lane elements include: lanes, intersections, and parking spaces.

[0074] In this embodiment, lane elements within the parking lot are obtained through neural network recognition based on the lane environment image. These lane elements include lanes, intersections, and parking spaces, and may also include other elements that can be used for coarse localization, such as speed bumps, signs, and pillars. The elements that must appear in the refined map are: lanes, intersections, and parking spaces. A lane is defined as a driving route between two intersections, and its direction needs to be distinguished. An intersection is a node that identifies the topological relationship between multiple roads. A parking space refers to the approximate location of the parking space frame and its ID.

[0075] Step S300: Based on the location information of the lanes and the intersections, construct a topological relationship diagram between each lane and the intersection.

[0076] Automatic cruise parking based on a high-precision map first requires obtaining a high-precision map, which in turn requires obtaining a topological relationship diagram between each lane element. In this embodiment, the topological relationship diagram is constructed based on the lanes and intersections in the lane elements and their corresponding location information, as well as the topological structure between each lane and each intersection.

[0077] Specifically, such as Figure 2 As shown, step S300 includes:

[0078] Step S301: The vehicle drives into the parking lot entrance and continuously collects lane environment images in the parking lot through the image acquisition device. The lane environment images are then used for image recognition to obtain the current perceived lane line.

[0079] In this embodiment, when a vehicle enters the parking lot entrance, the image acquisition device continuously acquires lane environment images within the parking lot, obtains lane environment images within the parking lot, and performs image recognition on the lane environment images based on the neural network, thereby identifying the current perceived lane line based on the lane environment images.

[0080] In step S302, the vehicle travels straight along the current perceived lane line. When the image recognizes an intersection, the perceived lane line between the parking lot entrance and the intersection is taken as the current lane.

[0081] In this embodiment, the vehicle travels straight along the current sensing lane line. When the image acquisition device captures and identifies an intersection, the intersection is taken as the first identified intersection, the sensing lane line between the parking lot entrance and the intersection is taken as the current lane, and the sensing lane line between the first intersection and the entrance is taken as the first lane.

[0082] In step S303, the vehicle travels straight along the next sensing lane line. When the image recognizes the next intersection, the next sensing lane line between the next intersection and the previous intersection is taken as the next lane, and the vehicle is controlled to continue driving and collect lane environment images until the parking lot exit is recognized.

[0083] In this embodiment, when the vehicle passes through the first intersection and continues straight along the next sensing lane line, when the image acquisition device identifies the next intersection, the next intersection is designated as the second intersection, and the lane line sensed between the first and second intersections is designated as the second lane. The identified lane elements at this time include: the first intersection, the second intersection, the first lane, and the second lane, as well as the topological relationship between these lane elements, i.e., the entrance is connected to the first intersection via the first lane, and the first intersection is connected to the second intersection via the second lane. The vehicle continues to drive and acquire lane environment images until the parking lot exit is identified, at which point all intersections and connecting lanes within the parking lot are identified.

[0084] Step S304: Obtain the location information of the lanes and intersections between the parking lot entrance and the parking lot exit, and construct a topological relationship diagram between each lane and the intersection.

[0085] In this embodiment, all lanes and intersections identified in step S303 are obtained, and the location information of the lanes and intersections is recorded. The topological relationship diagram between each lane and intersection is established by using the perceived lanes between the intersections identified above as topological connections.

[0086] This invention provides a schematic diagram of a process for constructing a topological relationship graph, as shown in the embodiment of the invention. Figure 3 As shown, specifically:

[0087] When the vehicle enters the parking lot entrance E, a sensing lane line is generated. The vehicle continues to drive along the currently sensed lane line. When the first intersection J1 is detected, the sensing lane line between the entrance E and the first intersection J1 is taken as the first lane L1. The sensing lane line continues to be generated after passing through the first intersection J0. When the second intersection J2 is detected, the sensing lane line between the first intersection J1 and the second intersection J2 is taken as the second lane L2. When the vehicle passes through the second intersection J2, the next intersection Jn is detected. The vehicle is controlled to continue driving and the lane environment image is collected until the parking lot exit is detected. A topology graph is constructed based on the detected first intersection J1, second intersection J2, first lane L1 and second lane L2.

[0088] Step S400: Based on the topology diagram, the parking space is attached to the corresponding lane according to the location information of the parking space, and a high-resolution map is generated.

[0089] In this embodiment, the construction of the high-precision map mainly consists of two parts: first, obtaining the topological relationship diagram between lane elements; and second, attaching parking spaces to the corresponding lanes, and generating the high-precision map based on the topological relationship diagram and the attached parking spaces. Step S300 has already completed the acquisition of each lane element and the topological relationship diagram between the lane elements. Step S400 mainly describes how to attach the parking spaces to the corresponding lanes based on the parking space information, and generate the high-precision map based on the attached lanes and the topological relationship diagram.

[0090] Specifically, such as Figure 4 As shown, step S400 includes:

[0091] Step S401: During vehicle operation, image recognition is performed on the real-time collected lane environment image to obtain parking space lines.

[0092] In this embodiment, during the vehicle's movement, the image acquisition device not only needs to acquire the lane environment, but also needs to acquire the parking spaces in the lane environment. The lane line corresponding to the parking space is the parking space line.

[0093] Step S402: Determine whether the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, based on the position information of the parking space line.

[0094] In this embodiment, the system determines whether a parking space exists in the current lane of the vehicle based on the location information of the parking space line. Specifically, the determination process is as follows:

[0095] If the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, and it is determined that the lane where the parking space is located is available for parking, then according to the location information of the parking space, the parking lot information and the topology diagram, the parking space is attached to the corresponding lane in the topology diagram. The parking lot information is the parking lot's unique ID, and the ID information of each parking lot is different, so no specific limitation is made here.

[0096] In this embodiment, if it is determined that the parking space corresponding to the parking space line exists in the current lane of the vehicle, then it is considered that the lane where the parking space is located can be used for parking. Therefore, the location information of the parking space can be attached to the topology graph.

[0097] If the parking space corresponding to the parking space line is not located in the current lane where the vehicle is traveling, it is determined that the lane where the parking space is located is not suitable for parking, and the lane will be discarded during this loading process.

[0098] In this embodiment, if it is determined that the parking space corresponding to the parking space line does not exist in the current lane where the vehicle is traveling, it is considered that the lane where the parking space is located cannot be used for parking. Therefore, the lane is discarded during the current cruise loading process, and it is determined in the next lane whether the parking space can be loaded into the next lane.

[0099] The present invention provides a schematic diagram of a process for mounting a parking space, as shown in the embodiment of the invention. Figure 5 As shown, specifically:

[0100] During vehicle operation, the perception module identifies a series of parking spaces such as P0, P1, and P2 (not shown in the diagram) in real time. Among the perceived parking spaces, it is determined that parking space P0 can be parked from the current lane L0, and parking space P1 can be parked through lane L1. However, parking space P2 cannot be parked through the currently perceived lane. Therefore, in this configuration, lanes L1 and L2 will be discarded for parking space P2. In the next configuration, it will be determined whether the position of parking space P2 exists in the current lane of the vehicle. If so, it is considered that the lane where the parking space is located can be parked, and therefore the position information of the parking space can be configured in the topology diagram.

[0101] After generating a high-resolution map in step S400 by attaching the parking space to the corresponding lane according to the topology diagram and the location information of the parking space, the method further includes, for parking lots that have not been entered for the first time, a high-resolution map of the parking lot has already been generated on the mobile terminal. In this embodiment, the existing high-resolution map is maintained and updated. Multi-vehicle data collection is essentially the addition, deletion, and modification of existing map elements. Specifically, the steps are as follows:

[0102] A preset high-precision map, wherein the preset high-precision map is a high-precision map corresponding to a known lane environment;

[0103] The vehicle's image acquisition device acquires lane environment images, and the lane environment images are compared with preset scene descriptors to determine whether the vehicle is driving on a lane in a preset high-precision map. The scene descriptor is the feature information of the lane environment in the preset high-precision map obtained through neural network learning.

[0104] If the vehicle is traveling in a lane in the preset standard definition map, then determine whether to update the parking spaces in the preset standard definition map based on the lane environment image;

[0105] If the vehicle is not traveling in a lane of the preset precision map, a new precision map is generated based on the lane environment image.

[0106] In this embodiment, the lane descriptor generated by perception is compared with the lane descriptor in the existing high-precision map to determine whether the vehicle is traveling on a known road. When the vehicle is traveling on a known road, the lane on the known high-precision map can be updated and loaded with parking spaces based on newly added or canceled parking spaces.

[0107] In another application scenario, when the vehicle approaches an intersection, the scene descriptor is compared with the descriptor of a known scene to determine if the current intersection location matches the intersection location in a known high-resolution map. If the similarity of the scene descriptors is higher than a preset threshold, it indicates that the current intersection location matches the intersection location in the known high-resolution map, and other road information is updated based on the road information connected to the intersection. If the similarity of the scene descriptors is lower than a preset threshold, it indicates that the current intersection location does not match the intersection location in the known high-resolution map, and the confidence of the original intersection is reduced. Once the confidence level drops to a certain threshold, the intersection is deleted, and a new intersection element is added.

[0108] Step S500: Based on the high-precision map, the vehicle automatically cruises and parks in the parking lot.

[0109] The specific S500 steps include:

[0110] The vehicle cruises based on the high-precision map;

[0111] Based on the parking spaces on the high-precision map, the system uses sensors to determine in real time whether the parking spaces are vacant.

[0112] If the parking space is vacant, then the parking space will be used as the target parking space for parking.

[0113] If the parking space is not vacant, the vehicle will cruise based on other parking spaces on the high-resolution map and their location information until it identifies a target parking space and parks there.

[0114] In this embodiment, the vehicle first cruises based on a high-precision map. By setting a parking space as the target parking space (i.e., the cruise endpoint), the vehicle plans a lane line for the cruise. Based on the location information of the parking space on the high-precision map, the vehicle uses sensing to determine in real time whether the parking space is vacant. If it is vacant, the vehicle uses the parking space as the target parking space for parking. If the parking space is not vacant, the vehicle resets the cruise endpoint based on the location information of other parking spaces on the high-precision map. The vehicle continues until it determines the target parking space during the cruise and then parks in the determined target parking space.

[0115] This invention provides a schematic diagram of a cruise process for mounting a parking space, as shown in the embodiment of the invention. Figure 6 As shown, specifically:

[0116] In this embodiment, SD Map cruise consists of the following steps: When entering from the entrance, the vehicle first uses scene descriptor matching to lock the current road L1 that the vehicle senses, searches the map to locate the road L3 where the target parking space is located, and finds a route from the current lane to the target lane through lane search. Cruise is carried out according to this process. At intersections, the vehicle can go straight or turn based on the topological relationship between roads. The cruise process relies on sensing lane lines and obstacles to keep the vehicle centered and avoid obstacles. After reaching the target road, the cruise ends and the vehicle uses perception to determine the position of the target parking space in real time and park there.

[0117] Specifically, in this embodiment, when a vehicle enters the parking lot from the entrance, the current lane is identified as L1, the target parking space is parking space 3, and the route is planned as L1-L2-L3. When the vehicle passes the first intersection j1, it determines that the next lane is L2, and at this time, it controls the vehicle to go straight. When it passes the second intersection j2, it determines that the next lane is L3, and at this time, it controls the vehicle to turn right. After reaching the target lane, the actual location of parking space 3 is found through scene recognition, and the straight-going automatic parking program parks the vehicle in parking space 3.

[0118] The present invention provides a schematic diagram of a process for mounting a parking space, as shown in the embodiment of the invention. Figure 7 As shown, by attaching the parking space to the obtained topology map according to the above steps S100-S400, a high-precision map can be obtained.

[0119] As can be seen from the above, in this invention, during vehicle operation, an image acquisition device installed on the vehicle acquires lane environment images within the parking lot; based on the lane environment images, the location information of lane elements within the parking lot is obtained, the lane elements including: lanes, intersections, and parking spaces; based on the location information of the lanes and intersections, a topological relationship diagram between each lane and the intersection is constructed; based on the topological relationship diagram, the parking space is attached to the corresponding lane according to the location information of the parking space, generating a high-precision map; based on the high-precision map, the vehicle automatically cruises and parks within the parking lot.

[0120] Compared with existing technologies, this invention provides an automated cruise parking method based on a high-precision map. The method obtains the location information of lane elements within the parking lot from lane environment images. These lane elements include lanes, intersections, and parking spaces. A topological relationship graph is constructed between the lanes and intersections. Based on the topological relationship graph, the location information of the parking spaces is mapped to the corresponding lanes to generate a high-precision map. The vehicle then automatically parks within the parking lot based on this high-precision map. This method of generating a high-precision map based on topological relationships does not rely on high-precision map positioning systems at the parking lot or vehicle end, which helps save on the cost of automated cruise parking and facilitates the large-scale deployment of automated cruise parking systems.

[0121] Exemplary device

[0122] This invention also provides an automatic cruise parking system based on a high-precision map, such as... Figure 8 As shown, the automatic cruise parking system based on a high-precision map includes:

[0123] The image acquisition module 610 is used to acquire lane environment images in a parking lot through an image acquisition device installed in the vehicle during vehicle operation;

[0124] The lane element acquisition module 620 is used to acquire the location information of lane elements in the parking lot based on the lane environment image. The lane elements include: lanes, intersections, and parking spaces.

[0125] The topology graph construction module 630 is used to construct a topology graph between each lane and the intersection based on the location information of the lanes and the intersections;

[0126] The high-precision map generation module 640 is used to generate a high-precision map by attaching the parking space to the corresponding lane according to the topology diagram and the location information of the parking space.

[0127] Automatic cruise parking module 650 is used for automatic cruise based on the high-precision map, and the vehicle parks in the parking lot.

[0128] The topology graph construction module 630 includes a current lane line acquisition unit 631, a current lane acquisition unit 632, a next lane acquisition unit 633, and a topology graph construction unit 634. Specifically, as shown in the figure... Figure 9 As shown.

[0129] The current perceived lane line acquisition unit 631 is used to continuously acquire lane environment images in the parking lot through an image acquisition device when the vehicle enters the parking lot entrance, and to perform image recognition on the lane environment images to obtain the current perceived lane line.

[0130] The current lane acquisition unit 632 is used for the vehicle to travel straight along the current perceived lane line. When the image recognizes an intersection, the perceived lane line between the parking lot entrance and the intersection is taken as the current lane.

[0131] The next lane acquisition unit 633 is used for the vehicle to travel straight along the next sensing lane line. When the image recognizes the next intersection, the next sensing lane line between the next intersection and the previous intersection is taken as the next lane, and the vehicle is controlled to continue to drive and collect lane environment images until the parking lot exit is recognized.

[0132] A relationship graph construction unit 634 is used to obtain the location information of the lanes and intersections between the parking lot entrance and the parking lot exit, and to construct a topological relationship graph between each lane and the intersection.

[0133] It should be noted that the division of the various modules of the automatic cruise parking system based on the high-precision map is not unique and is not intended as a specific limitation.

[0134] Based on the above embodiments, the present invention also provides a smart terminal, the principle block diagram of which can be as follows: Figure 10 As shown. The intelligent terminal includes a processor and a memory. The memory of the intelligent terminal includes an automatic cruise parking program based on a high-precision map, and the memory provides an environment for the operation of the automatic cruise parking program based on the high-precision map. When the processor executes the automatic cruise parking program based on the high-precision map, it implements the steps of any of the automatic cruise parking methods based on the high-precision map. It should be noted that the intelligent terminal may also include other functional modules or units, which are not specifically limited here.

[0135] Those skilled in the art will understand that Figure 10 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the smart terminal to which the present invention is applied. Specifically, the smart terminal may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0136] This invention also provides a computer-readable storage medium storing an automatic cruise parking program based on a high-precision map. When executed by a processor, the automatic cruise parking program based on the high-precision map implements the steps of any of the automatic cruise parking methods based on high-precision maps provided in this invention.

[0137] It should be understood that the sequence number of each step in the embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0138] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the division of the functional units and modules is only described as an example. In practical applications, the functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0139] In the embodiments described, each embodiment has its own emphasis. For parts not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0140] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0141] In the embodiments provided by this invention, it should be understood that the disclosed systems / smart terminals and methods can be implemented in other ways. For example, the system / smart terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, it can be done in other ways. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0142] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0143] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not mean that the essence of the corresponding technical solutions deviates from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An automatic cruise parking method based on a high-precision map, characterized in that, include: During vehicle operation, images of the lane environment within the parking lot are captured by image acquisition devices installed on the vehicle. The location information of lane elements in the parking lot is obtained based on the lane environment image. The lane elements include: lane, intersection and parking space. The lane is defined as a driving route between two intersections. The lane also includes directions. The intersection is used to identify nodes that identify the topological relationship between multiple roads. The parking space is used to provide the location of the parking space frame and the parking space ID. Based on the location information of the lanes and the intersections, a topological relationship diagram between each lane and the intersection is constructed; The step of constructing a topological relationship graph between each lane and each intersection based on the location information of the lanes and intersections includes: As the vehicle enters the parking lot entrance, it continuously collects lane environment images within the parking lot through image acquisition equipment, performs image recognition on the lane environment images, and obtains the currently perceived lane lines. The vehicle travels straight along the current perceived lane line. When the image recognizes an intersection, the perceived lane line between the parking lot entrance and the intersection is taken as the current lane. Based on the topology diagram, the parking spaces are attached to the corresponding lanes according to their location information to generate a high-resolution map. Based on the precise map, the vehicle automatically cruises and parks itself in the parking lot. Based on the topology diagram, the parking spaces are mapped to the corresponding lanes according to their location information, generating a high-resolution map, including: During vehicle operation, image recognition is performed on the real-time collected lane environment images to obtain parking space lines; Based on the location information of the parking space line, determine whether the parking space corresponding to the parking space line is set in the current lane where the vehicle is traveling; When the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, the parking space is attached to the corresponding lane in the topology map according to the location information of the parking space, and a high-precision map is generated.

2. The method according to claim 1, characterized in that, The step of constructing a topological relationship graph between each lane and each intersection based on the location information of the lanes and intersections further includes: The vehicle travels straight along the next sensing lane line. When the image recognizes the next intersection, the next sensing lane line between the next intersection and the previous intersection is taken as the next lane, and the vehicle is controlled to continue driving and collect lane environment images until the parking lot exit is recognized. Obtain the location information of the lanes and intersections between the parking lot entrance and the parking lot exit, and construct a topological relationship diagram between each lane and the intersection.

3. The method according to claim 1, characterized in that, The step of determining whether the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, based on the position information of the parking space line, includes: If the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, and it is determined that the lane where the parking space is located is available for parking, then according to the location information of the parking space, the parking lot information and the topology diagram, the parking space is attached to the lane corresponding to the topology diagram; If the parking space corresponding to the parking space line is not located in the current lane where the vehicle is traveling, it is determined that the lane where the parking space is located is not suitable for parking, and the lane will be discarded during this loading process.

4. The method according to claim 1, characterized in that, After generating a high-resolution map by attaching the parking space to the corresponding lane based on the location information of the parking space according to the topology graph, the method further includes: A preset high-precision map, wherein the preset high-precision map is a high-precision map corresponding to a known lane environment; The vehicle's image acquisition device acquires lane environment images, and the lane environment images are compared with preset scene descriptors to determine whether the vehicle is driving on a lane in a preset high-precision map. The scene descriptor is the feature information of the lane environment in the preset high-precision map obtained through neural network learning. If the vehicle is driving in a lane in a preset high-precision map, then determine whether to update the parking spaces in the preset high-precision map based on the lane environment image. If the vehicle is not traveling in a lane of the preset precision map, a new precision map is generated based on the lane environment image.

5. The method according to claim 1, characterized in that, The automatic cruise based on the high-precision map, where the vehicle parks within the parking lot, includes: The vehicle cruises based on the high-precision map; Based on the parking spaces on the high-precision map, the system uses sensors to determine in real time whether the parking spaces are vacant. If the parking space is vacant, then the parking space will be used as the target parking space for parking. If the parking space is not vacant, the vehicle will cruise based on other parking spaces on the high-resolution map and their location information until it identifies a target parking space and parks there.

6. An automatic cruise parking system based on a high-precision map, characterized in that, The system includes: The image acquisition module is used to acquire images of the lane environment in the parking lot through image acquisition devices installed in the vehicle during vehicle operation; The lane element acquisition module is used to acquire the location information of lane elements in the parking lot based on the lane environment image. The lane elements include: lane, intersection and parking space. The lane is defined as a driving route between two intersections. The lane also includes directions. The intersection is used to identify nodes that identify the topological relationship between multiple roads. The parking space is used to provide the location of the parking space frame and the parking space ID. The topology graph construction module is used to construct a topology graph between each lane and the intersection based on the location information of the lanes and the intersections; The topology graph construction module includes: The current perceived lane line acquisition unit is used when the vehicle enters the parking lot entrance, continuously acquires lane environment images in the parking lot through image acquisition equipment, performs image recognition on the lane environment images, and obtains the current perceived lane line. The current lane acquisition unit is used for the vehicle to travel straight along the current perceived lane line. When the image recognizes an intersection, the perceived lane line between the parking lot entrance and the intersection is taken as the current lane. The high-precision map generation module is used to attach the parking space to the corresponding lane according to the topology diagram and the location information of the parking space to generate a high-precision map; An automatic cruise parking module is used for automatic cruise based on the high-precision map, in which the vehicle parks itself in the parking lot. Based on the topology diagram, the parking spaces are mapped to the corresponding lanes according to their location information, generating a high-resolution map, including: During vehicle operation, image recognition is performed on the real-time collected lane environment images to obtain parking space lines; Based on the location information of the parking space line, determine whether the parking space corresponding to the parking space line is set in the current lane where the vehicle is traveling; When the parking space corresponding to the parking space line is located in the current lane where the vehicle is traveling, the parking space is attached to the corresponding lane in the topology map according to the location information of the parking space, and a high-precision map is generated.

7. The automatic cruise parking system based on a high-precision map according to claim 6, characterized in that, The topology graph construction module also includes: The next lane acquisition unit is used for the vehicle to travel straight along the next sensing lane line. When the image recognizes the next intersection, the next sensing lane line between the next intersection and the previous intersection is taken as the next lane, and the vehicle is controlled to continue to drive and collect lane environment images until the parking lot exit is recognized. A relationship graph construction unit is used to obtain the location information of the lanes and intersections between the parking lot entrance and the parking lot exit, and to construct a topological relationship graph between each lane and the intersection.

8. A smart terminal, characterized in that, The smart terminal includes a memory, a processor, and an automatic cruise parking program based on a high-precision map stored in the memory and executable on the processor. When the automatic cruise parking program based on the high-precision map is executed by the processor, it implements the steps of the automatic cruise parking method based on a high-precision map as described in any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an automatic cruise parking program based on a high-precision map, which, when executed by a processor, implements the steps of the automatic cruise parking method based on a high-precision map as described in any one of claims 1-5.

Images