Valet parking service method and system

Abstract

The present invention provides a valet parking service method, which is used for realizing a valet parking service system for automatic parking of vehicles, and includes: receiving one or more parking requests and post-parking requirements initiated by vehicles; calculating the priority data of each vehicle entering the corresponding target parking lot according to the parking requests and post-parking requirements initiated for each vehicle, and combining the first parameter combination data fed back by each target parking lot to obtain the user alighting point coordinates of each vehicle and the corresponding allocated vacant parking space, so as to plan the parking entry path of each vehicle; when it is detected that each vehicle has arrived at the corresponding user alighting point coordinates and the user has alighted, based on the parking entry path of each vehicle, controlling each vehicle to automatically park into the corresponding allocated vacant parking space. The implementation of the present invention can solve the problems of a single request data source, parking space allocation not meeting user expectations, and parking congestion existing in the existing valet parking method.

Description

Technical Field

[0001] The present invention relates to the field of automobile technology, and in particular to a valet parking service method and system. Background Art

[0002] The term "valet parking" comes from the service provided by hotels and other places for customers to park their cars after the key is handed over. Valet parking in autonomous driving, as the name suggests, refers to the function of the vehicle to park in the target parking space after autonomously completing path planning, self-positioning, and obstacle avoidance control in the absence of anyone. At present, it has become an industry consensus to achieve valet parking based on end-to-end deep integration. With the support of 5G communications, the Internet of Things, high-precision maps, GNSS&IMU fusion navigation, visual SLAM and other technologies, the goal of valet parking has gradually shifted from accurate and safe parking in parking spaces to providing users with comfortable and convenient parking services. Therefore, relying on autonomous mapping, autonomous navigation, autonomous positioning and autonomous obstacle avoidance technologies accurate to the centimeter level, it is possible to park accurately and safely in a parking space.

[0003] However, the existing valet parking methods have the following shortcomings, which are: (1) the source of user parking request data is single and does not include the user's other parking intentions; for example, shopping, dining, working, etc.; (2) the factors considered when allocating vacant parking spaces are not comprehensive, and the actual needs of users after parking are not considered, resulting in parking space allocation that does not meet user expectations; for example, shopping, dining, working, etc., it is easy for the allocated parking space to be too far from the user's actual destination, resulting in poor valet parking performance; (3) when the user's drop-off point is fixed, parking congestion is not considered; for example, the user's drop-off point is usually fixed at the entrance of the parking lot, and congestion is more serious during peak parking hours.

[0004] Therefore, it is necessary to design a new valet parking service method, which can solve the above-mentioned problems existing in the existing valet parking method and provide users with high-quality parking services. Summary of the invention

[0005] The technical problem to be solved by the embodiments of the present invention is to provide a valet parking service method and system, which can solve the problems of a single request data source, parking space allocation not meeting user expectations, and parking congestion in the existing valet parking method, and can provide users with high-quality parking services.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a valet parking service method, which is used in a valet parking service system for realizing automatic parking of a vehicle. The method comprises the following steps:

[0007] Receiving one or more parking requests and post-parking requirements initiated by a vehicle;

[0008] According to the parking request initiated by each vehicle and the demand after parking, the priority data of each vehicle entering the corresponding target parking lot is calculated, and the coordinates of the user's alighting point of each vehicle in the corresponding target parking lot and the corresponding allocated vacant parking space are obtained in combination with the first parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot, and the parking entry path of each vehicle from the coordinates of the user's alighting point to the allocated vacant parking space in the corresponding target parking lot is further planned;

[0009] When it is detected that each vehicle has arrived at the coordinates of the user's getting-off point in the corresponding target parking lot and the user has got off the vehicle, each vehicle is controlled to automatically park into the corresponding allocated vacant parking space based on the parking entry path planned for each vehicle.

[0010] Wherein, the method further comprises:

[0011] According to the parking request initiated by each vehicle and the demand after parking, the priority data of each vehicle driving out of the corresponding target parking lot is calculated, and the coordinates of the user boarding point of each vehicle in the corresponding target parking lot are obtained by combining the second parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot, and further combined with the parking space position of each vehicle in the corresponding target parking lot, the parking exit path of each vehicle from the parking space position to the user boarding point coordinates in the corresponding target parking lot is planned;

[0012] Based on the parking exit paths planned for each vehicle, each vehicle is controlled to automatically park and exit to the corresponding user boarding point coordinates.

[0013] Among them, the mapping relationships pre-constructed based on the key coordinate points of each target parking lot include: the mapping relationship between the simple diagram information of the key coordinate points of the parking lot and the resource information around the parking lot, the mapping relationship between the simple diagram information of the key coordinate points and the parking space location information, and the mapping relationship between the key coordinate points of the parking lot and the get-on and get-off constraints of the parking lot roads.

[0014] Among them, the coordinates of the user alighting point and the user boarding point of each vehicle are based on the parking needs initiated by each vehicle, and the positioning conditions are the minimum walking time for pedestrians and the minimum time for vehicles to enter or exit the associated parking spaces. The key coordinate points of each vehicle in the corresponding target parking lot are first located, and then the coordinates are determined by calculating the minimum pedestrian walking time from the located key coordinate points.

[0015] Wherein, the method further comprises:

[0016] When controlling each vehicle to perform automatic parking, comparing in real time whether the actual driving path of each vehicle is consistent with the corresponding planned parking entry path or parking exit path; and,

[0017] If one or more vehicles report that there are obstacles on the actual driving path that cannot be bypassed, a request is made to re-plan the routes for all vehicles with the warning monitoring conditions, and all vehicles with the warning monitoring conditions are further controlled to park in or out according to their respective re-planned routes.

[0018] Among them, the actual driving path of each vehicle is obtained through SLAM mapping matching combined with wheel speed and IMU fusion positioning technology; the early warning monitoring situation is detected through vision and radar fusion technology.

[0019] The priority data of each vehicle entering the corresponding target parking lot includes the time point of receiving the parking request initiated by each vehicle and the parking demand, and the distance from the current position of each vehicle to the corresponding target parking lot;

[0020] The priority data for each vehicle to exit the corresponding target parking lot includes the time point at which the parking request and the parking demand initiated for each vehicle are received, and the distance from the current position of each vehicle to the user's expected boarding point in the corresponding target parking lot.

[0021] The parking demand may be directly input by the user, or automatically determined based on the type of destination input by the user, or actively recommended based on parking lots near the destination input by the user and determined by the user again.

[0022] The embodiment of the present invention further provides a valet parking service system, comprising: a smart mobile terminal, a travel service provider and a valet parking controller; wherein:

[0023] There is at least one smart mobile terminal connected to the travel service provider; each smart mobile terminal is used to receive a parking request and a post-parking demand initiated by a user for the vehicle being driven and riding, and report the request to the travel service provider;

[0024] The travel service provider includes a parking entry path planning unit; wherein the parking entry path planning unit is used to calculate the priority data of each vehicle entering the corresponding target parking lot according to the parking request initiated by each vehicle and the demand after parking, and obtain the user alighting point coordinates of each vehicle in the corresponding target parking lot and the corresponding allocated vacant parking space in combination with the first parameter combination data fed back by the mapping relationship pre-constructed based on the key coordinate points of each target parking lot, and plan the parking entry path of each vehicle from the user alighting point coordinates to the allocated vacant parking space in the corresponding target parking lot, and further send it to the valet parking controller;

[0025] There is at least one valet parking controller, which is pre-installed on a vehicle driven by a corresponding user and then connected to the travel service provider; wherein each valet parking controller includes a parking entry control unit, which is used to control the vehicle to automatically park into the corresponding allocated vacant parking space based on the corresponding planned parking entry path of the vehicle when it detects that the vehicle has arrived at the user get-off point coordinates in the corresponding target parking lot and the user has got off the vehicle.

[0026] The travel service provider further includes a parking exit path planning unit; wherein the parking exit path planning unit is used to calculate the priority data of each vehicle exiting the corresponding target parking lot according to the parking request initiated for each vehicle and the demand after parking, and to obtain the coordinates of the user boarding point of each vehicle in the corresponding target parking lot in combination with the second parameter combination data fed back by the mapping relationship pre-constructed based on the key coordinate points of each target parking lot, and to plan the parking exit path of each vehicle from the parking space position to the user boarding point coordinates in the corresponding target parking lot in combination with the parking space position of each vehicle in the corresponding target parking lot, and further send it to the valet parking controller;

[0027] Each valet parking controller further includes a parking exit control unit, which is used to control the vehicle to automatically park and exit to the corresponding user boarding point coordinates based on the parking exit path corresponding to the vehicle.

[0028] Implementing the embodiments of the present invention has the following beneficial effects:

[0029] 1. The present invention adds a user's parking demand selection option based on the original collection of the user's intended parking location, collects the user's actual parking demand, and thus solves the problem of a single request data source in the existing valet parking method;

[0030] 2. In view of the problem that the allocation of free parking spaces is not in line with user expectations because only three factors, namely, the user queue priority, the distance between the free parking spaces and the driver's getting off point, and the presence or absence of obstacles between the free parking spaces and the driver's getting off point, are considered when allocating free parking spaces, this invention collects the first parameter combination data fed back by the mapping relationship pre-built by the target parking lot based on the key coordinate points (such as the mapping relationship between the resource information around the parking lot, the parking space location information, and the parking and getting off constraints on the parking lot road), and combines the collected user's parking needs to reasonably allocate free parking spaces through an artificial intelligence algorithm, which is more in line with the user's actual use needs, thereby solving the problem that the parking space allocation of the existing valet parking method does not meet the user's expectations, and can provide users with high-quality parking services;

[0031] 3. To address the parking congestion problem caused by fixed user alighting points, the present invention collects user parking needs, comprehensively considers the first parameter combination data fed back by the mapping relationship pre-built by the target parking lot based on key coordinate points (such as the mapping relationship between the resource information around the parking lot, the parking space location information, and the parking lot road alighting constraints), and obtains the dynamic user alighting point coordinates after comprehensive calculation, which greatly alleviates parking congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying creative labor, other drawings obtained based on these drawings still belong to the scope of the present invention.

[0033] Figure 1 A schematic diagram of the structure of a valet parking service system provided by an embodiment of the present invention;

[0034] Figure 2 The present invention provides a flowchart of a valet parking service method. DETAILED DESCRIPTION

[0035] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0036] like Figure 1 As shown, a valet parking service system is provided in an embodiment of the present invention, including: a smart mobile terminal 1, a travel service provider 2 and a valet parking controller 3.

[0037] At this time, there is at least one smart mobile terminal 1, which is connected to the travel service provider 2; each smart mobile terminal 1 is used to receive parking requests and post-parking requirements initiated by users for their vehicles, and report them to the travel service provider 2.

[0038] At this time, the travel service provider 2 includes a parking entry path planning unit and a parking exit path planning unit; wherein,

[0039] The parking entry path planning unit is used to calculate the priority data of each vehicle entering the corresponding target parking lot according to the parking request initiated for each vehicle and the demand after parking, and obtain the coordinates of the user alighting point of each vehicle in the corresponding target parking lot and the corresponding allocated vacant parking space in combination with the first parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot, and plan the parking entry path of each vehicle from the coordinates of the user alighting point to the allocated vacant parking space in the corresponding target parking lot, and further send it to the valet parking controller 3; wherein, the priority data of each vehicle entering the corresponding target parking lot includes the time point of receiving the parking request initiated for each vehicle and the demand after parking, and the distance from the current position (i.e., the current driving position) of each vehicle to the corresponding target parking lot;

[0040] The parking exit path planning unit is used to calculate the priority data of each vehicle exiting the corresponding target parking lot according to the parking request and the demand after parking initiated to each vehicle, and obtain the coordinates of the user boarding point of each vehicle in the corresponding target parking lot in combination with the second parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot, and plan the parking exit path of each vehicle from the parking space position to the coordinates of the user boarding point in the corresponding target parking lot in combination with the parking space position of each vehicle in the corresponding target parking lot, and further send it to the valet parking controller 3. Among them, the priority data of each vehicle exiting the corresponding target parking lot includes the time point of receiving the parking request and the demand after parking initiated to each vehicle, and the distance from the current position of each vehicle (i.e., the parking space where the vehicle is located) to the expected user boarding point in the corresponding target parking lot.

[0041] It should be noted that the mapping relationship pre-built based on the key coordinate points of each target parking lot is set on the parking lot management terminal that communicates with the travel service provider 2, specifically including: the mapping relationship between the parking lot key coordinate point diagram information and the resource information around the parking lot, the mapping relationship between the key coordinate point diagram information and the parking space location information, and the mapping relationship between the parking lot key coordinate points and the parking lot road get-on and get-off constraints. At this time, the user get-off point coordinates and user get-on point coordinates of each vehicle are based on the post-parking demand initiated by each vehicle, and the minimum pedestrian walking time and the minimum time for the vehicle to enter or exit the associated parking space are used as positioning conditions. The key coordinate points of each vehicle in the corresponding target parking lot are first located, and then determined by calculating the minimum pedestrian walking time from the located key coordinate points. For example, a vehicle initiates a parking demand at Supermarket A for shopping. The parking lot of Supermarket A is located, and the walking time for pedestrians to enter Supermarket A from the parking lot T1 entrance and the time for vehicles to drive into an empty parking space from the T1 entrance are found to be the minimum. Then, the T1 entrance is located as the key coordinate point, and the coordinate point with the minimum walking time for pedestrians from the T1 entrance is calculated as the user alighting point of the vehicle. For another example, a vehicle initiates a parking demand at Supermarket A to go home. The parking lot of Supermarket A is located, and the walking time for pedestrians to enter the T2 elevator exit of the parking lot from the A supermarket is found to be the minimum and the time for vehicles to drive into the T2 elevator exit from the parking space is found to be the minimum. Then, the T2 elevator exit is located as the key coordinate point, and the coordinate point with the minimum walking time for pedestrians from the T2 elevator exit is calculated as the user boarding point of the vehicle.

[0042] At this time, there is at least one valet parking controller 3, which is pre-installed on the vehicle driven by a corresponding user and then connected to the travel service provider 2; wherein,

[0043] Each valet parking controller 3 includes a parking entry control unit and a parking exit control unit; the parking entry control unit is used to control each vehicle to automatically park into the corresponding allocated vacant parking space based on the parking entry path corresponding to the vehicle when it detects that the vehicle has arrived at the user's getting-off point coordinates in the corresponding target parking lot and the user has got off the vehicle; the parking exit control unit is used to control the vehicle to automatically park out to the corresponding user's getting-on point coordinates based on the parking exit path corresponding to the vehicle.

[0044] It should be noted that when the valet parking controller 3 controls each vehicle to perform automatic parking (driving in or out), it obtains the actual driving path of each vehicle through SLAM mapping matching and combines the wheel speed and IMU fusion positioning technology, and compares the actual driving path of each vehicle in real time to see whether it is consistent with the corresponding planned parking entry path or parking exit path.

[0045] At the same time, through the technology of integrating vision and radar, it is detected whether there are obstacles that cannot be bypassed on the actual driving path of each vehicle, and when an obstacle that cannot be bypassed is detected, the warning monitoring situation is fed back to the valet parking controller 3. Therefore, if a valet parking controller 3 receives a warning monitoring situation from the corresponding vehicle that there are obstacles that cannot be bypassed on the actual driving path, it requests the travel service provider 2 to re-plan the path of the vehicle, and after waiting for the re-planned path issued by the travel service provider 2 upon request, it continues to control the vehicle to park in or out according to the re-planned path.

[0046] In the embodiment of the present invention, the intelligent mobile terminal 1 includes a smart phone and a car computer, which provides an interactive entrance for the user to select and start the valet parking service. Among them, the parking demand can be obtained in the following three ways: (1) the user actively enters the parking demand in the valet parking APP, that is, the user directly enters it; (2) after the user enters the destination in the valet parking APP, the APP automatically judges the demand according to the user's destination type (restaurant, cinema, supermarket, etc.), that is, the demand is automatically judged according to the type of destination entered by the user; (3) after the user enters the parking lot near the destination in the valet parking APP, the APP actively recommends the destination and the user confirms it, that is, the parking lot near the destination entered by the user is actively recommended and the demand is judged after the user confirms it again.

[0047] In an embodiment of the present invention, when the travel service provider 2 implements the allocation of free parking spaces, dynamic alighting point allocation and parking alighting path planning through the parking entry path planning unit, the time point when the parking request initiated by the vehicle and the demand after parking (i.e., the parking signal) are received is used as the user sorting priority parameter A, the distance from the current position of the vehicle (i.e., the current driving position) to the corresponding target parking lot is used as the user sorting priority parameter B, the degree of fit between the user's destination and the resource information around the parking lot is used as the input parameter C, the mapping relationship between the resource information around the parking lot and the simplified diagram information of the key coordinate points of the parking lot is used as the input parameter D, the time point when the user actually arrives near the parking point is used as the input parameter E, the mapping relationship between the current empty parking space and the simplified diagram of the key coordinate points of the parking lot is used as the input parameter F, and the dynamic boarding / disembarking point array near the key coordinate points is used as the input parameter G. The above parameters are compiled into an array [A, B, C, D, E, F, G] uniquely associated with the user ID as the input of the artificial intelligence algorithm. The output of the artificial intelligence algorithm is the coordinates of the planned user drop-off point in the parking lot {X1, Y1, Z1}, the assigned parking space number XXXX, and the parking entry path Path_Hope1[X][Y][Z] from the user drop-off point to the assigned parking space. Since the input parameters will change with the actual position of the vehicle, the user's final intention, and the situation of SLAM detection obstacles, the output of the artificial intelligence will also be updated in real time.

[0048] When the travel service provider 2 implements dynamic pick-up point allocation and parking exit path planning through the parking exit path planning unit, the time point of receiving the parking request initiated by the vehicle and the demand after parking (i.e., the parking signal) is used as the user sorting priority parameter A, the distance between the current parking space of the vehicle and the user's expected pick-up point is used as the user sorting priority parameter B, the mapping relationship between the resource information around the parking lot and the simplified diagram information of the key coordinate points of the parking lot is used as the input parameter C, the mapping relationship between the simplified diagram information of the key coordinate points of the parking lot and the location information of the parking space is used as the input parameter D, and the array of dynamic pick-up / drop-off points near the key coordinate points is used as the input parameter E. The above parameters are compiled into an array [A, B, C, D, E] uniquely associated with the user ID as the input of the artificial intelligence algorithm, and the output of the artificial intelligence algorithm is the parking exit path Path_Hope2[X][Y][Z] from the parking space to the user pick-up point {X2, Y2, Z2}. Since the input parameters will change with the actual position of the vehicle, the final intention of the user, and the obstacle situation detected by SLAM, the output of the artificial intelligence will also be updated in real time. It should be noted that the user's expected boarding point is the initial user boarding point automatically given based on the demand after parking, combined with the resource information around the parking lot, the schematic information of the key coordinate points of the parking lot and the parking space location information.

[0049] It is understandable that the travel service provider 2 will also push the parking entry data such as the user's alighting point coordinates, allocated vacant parking spaces and parking entry paths of each vehicle in the corresponding target parking lot, as well as the parking exit data such as the user's boarding point coordinates, parking space locations and parking exit paths of each vehicle in the corresponding target parking lot to the corresponding smart mobile terminal 1 for display, so as to show the user the expected valet parking results.

[0050] In the embodiment of the present invention, when the valet parking controller 3 controls the vehicle to perform automatic parking through the parking entry control unit, the vehicle drives along the parking entry path Path_Hope1[X][Y][Z] from the user's alighting point to the allocated parking space and finally parks in the parking space XXXX, specifically: downloading the planned path and parking lot key diagram information from the travel service provider 2, detecting that the user drives into the electronic fence, identifying the key elements in the surrounding environment (navigation signs, parking space numbers, etc.) through SLAM mapping and matching them with the parking lot key diagram information, after the match is successful, the self-positioning is realized through the wheel speed and IMU fusion positioning and real-time matching is performed with the planned parking entry path starting point (i.e., the user's alighting point), and after detecting that the vehicle is about to drive to the user's alighting point, the user is prompted to get off through the smart mobile terminal 1. After the user gets off the vehicle, the vehicle control is started to enter the valet parking cruise stage. During the cruise stage, the valet parking controller continuously determines the current vehicle driving path Path_Now[X][Y][Z] through SLAM mapping matching, wheel speed and IMU fusion positioning, and compares it with the planned parking entry path Path_Hope1[X][Y][Z] in real time. If no obstacles that cannot be bypassed are detected by means of vision & radar fusion during the cruise stage, the vehicle is controlled to continue driving along the parking entry path Path_Hope1[X][Y][Z] until it is parked in a pre-assigned vacant parking space; conversely, if obstacles that cannot be bypassed are detected by means of vision & radar fusion during the cruise stage, the travel service provider 2 is requested to re-plan the path and allocate parking spaces, and wait for the travel service provider 2 to update the results after receiving the request, and then re-control the vehicle for automatic parking entry.

[0051] When the valet parking controller 3 controls the vehicle to perform automatic parking through the parking exit control unit, the vehicle exits the parking space and arrives at the user boarding point according to the parking exit path Path_Hope2[X][Y][Z] from the parking space to the user boarding point. Specifically, the planned path and parking lot key diagram information are downloaded from the travel service provider 2. After the download is completed, the vehicle is started to enter the cruising stage, and the current vehicle position and parking exit path Path_Now[X][Y][Z] are continuously determined through SLAM map matching, wheel speed and IMU fusion positioning, and compared with the planned parking exit path Path_Hope2[X][Y][Z] in real time. If no insurmountable obstacle is detected by means of vision & radar fusion during the cruising stage, the vehicle is controlled to continue to travel along the parking exit path Path_Hope2[X][Y][Z] until the coordinates {X2, Y2, Z2} pre-assigned to the user boarding point are reached; conversely, if an insurmountable obstacle is detected by means of vision & radar fusion during the cruising stage, the travel service provider 2 is requested to replan the path, and the travel service provider 2 waits for the updated result after receiving the request, and then controls the vehicle again to automatically park and exit.

[0052] In the embodiment of the present invention, the parking lot management terminal communicates with the travel service provider 2, and pre-defines a variety of mapping relationships based on key coordinate points, specifically:

[0053] (1) Establish a mapping relationship between resource information around the parking lot and the key coordinate point diagram information of the parking lot; divide the existing resources around the parking lot into N types {Entertainment, Restaurant, Education, Supermarket, etc.}, each resource type is an array with M elements in the array; divide the key coordinate points in the key coordinate point diagram information of the parking lot into K types {Elevator, Stairs, Gate, etc.}, each resource type is an array with J elements in the array; establish a mapping relationship between the key coordinate point type elements and the resource type elements one by one, and there are a total of (N*M)*(K*J) types of mappings. The time P_Walk_Time for pedestrians to complete walking from the key node of the parking lot to the resource point in each mapping is used as the sorting priority of each mapping. After the sorting is completed, the mapping relationship is established.

[0054] (2) Establish a mapping relationship between the key coordinate point diagram information of the parking lot and the parking space location information: Establish the Park_Lot[P] array with all the parking space location information of the parking lot, and divide the key coordinate points in the key coordinate point diagram information of the parking lot into K types {Elevator, Stairs, Gate,...}. Each resource type is an array with J elements in the array. A mapping relationship is established one by one between the parking space location elements and the key coordinate point type elements. There are a total of P*(K*J) types of mappings. The time AutoD_Time for a vehicle to complete autonomous driving from a key node of the parking lot to a parking space under each mapping is used as the sorting priority of each mapping. After the sorting is completed, the mapping relationship is established.

[0055] (3) Establishing a dynamic array of boarding / alighting points near key coordinate points: Obtain the coordinates of the key coordinate points and the parking constraints of the parking lot road. Take a boarding / alighting area of ​​6 m in length and 2.6 m in width. Cut the area near the key coordinate points within the range of the road parking constraints. After cutting, use the center coordinate point of each boarding / alighting area as an element to form a series of array structures Elevator.Park_Point[X][Y] / Stairs.Park_Point[X][Y], etc. The array is established by arranging the elements in each array structure according to the distance from the key point coordinates.

[0056] Taking two vehicles as an example, an application scenario of a valet parking service system in an embodiment of the present invention is further described as follows:

[0057] First, user User_A enters the supermarket Supermarket_A near the parking lot Parking_Space in the valet parking APP on a smart mobile terminal 1. The APP then actively recommends the destination parking lot Parking_Space for valet parking. After the user confirms the demand, the APP packages the user demand {User_A.Supermarket_A, User_A.Now_Location, User_A.Parking_Space, User_A.Send_Req_Time} and sends it to the travel service provider 2.

[0058] At this time, the mapping relationship between the resource information around the parking lot provided by the parking lot management terminal and the simplified diagram information of the key coordinate points of the parking lot shows that the pedestrian walking time P_Walk_Time value between the resource Parking_Space.Supermarket_A and the key coordinate point Parking_Space.Gate[1] is the smallest.

[0059] The mapping relationship between the key coordinate point diagram information and the parking space location information provided by the parking lot management terminal shows that the vehicle travel time AutoD_Time value of the key coordinate points Parking_Space.Gate[1] and Park_Lot[1] is the smallest.

[0060] The dynamic array of boarding / dropping points near the key coordinate points provided by the parking management terminal shows that the parking point Gate[1].Park_Point[X][Y] is closest to the key coordinate point Parking_Space.Gate[1].

[0061] User User_B enters Restaurant_B near parking lot Parking_Space in the valet parking APP on another smart mobile terminal 1. The APP then actively recommends the destination parking lot Parking_Space for valet parking. After the user confirms the demand, the APP packages the user demand {User_B.Restaurant_B,User_B.Now_Location,User_B.Parking_Space,User_A.Send_Req_Time} and sends it to travel service provider 2.

[0062] At this time, the mapping relationship between the resource information around the parking lot provided by the parking lot management terminal and the simplified diagram information of the key coordinate points of the parking lot shows that the pedestrian walking time P_Walk_Time value between the resource Parking_Space.Restaurant_B and the key coordinate point Parking_Space.Elevator[2] is the smallest.

[0063] The mapping relationship between the key coordinate point diagram information and the parking space location information provided by the parking lot management terminal shows that the vehicle travel time AutoD_Time value of the key coordinate points Parking_Space.Elevator[2] and Park_Lot[2] is the smallest.

[0064] The dynamic array of boarding / dropping points near the key coordinate points provided by the parking management terminal shows that the parking point Elevator[2].Park_Point[X][Y] is closest to the key coordinate point Parking_Space.Elevator[2].

[0065] Secondly, after receiving the above data, travel service provider 2 starts to calculate and plan the driving path between Parking_Space.Gate[1] and Park_Lot[1], assigns parking space Park_Lot[1] to user User_A, and assigns parking point Gate[1].Park_Point[X][Y] to User_A. It starts to calculate and plan the driving path between Parking_Space.Elevator[2] and Park_Lot[2], assigns parking space Park_Lot[2] to user User_B, and assigns parking point Elevator[2].Park_Point[X][Y] to User_B.

[0066] Next, after completing the calculation, the travel service provider 2 sends the calculation results to the smart mobile terminals 1 of User_A and User_B respectively, and User_A and User_B confirm that the calculation results are valid.

[0067] Next, User_A and User_B set off in their cars, and Travel Service Provider 2 monitors the location of the vehicles at all times. User_A generates a User_A_Arrive_Time parameter when it arrives near the target parking lot, and User_B generates a User_B_Arrive_Time parameter when it arrives near the target parking lot. Travel Service Provider 2 sorts the User_A_Arrive_Time parameter and the User_B_Arrive_Time parameter in their respective sequences. The sorting results show that User_A_Arrive_Time and User_B_Arrive_Time are both ranked high and do not need to be re-planned.

[0068] Finally, User_A drives into the electronic fence, and the valet parking controller 3 on his vehicle shows a successful match. User_A continues to drive to the vicinity of Gate[1].Park_Point[X][Y]. The valet parking controller 3 on the vehicle reminds the user to get off. After the user gets off, the valet parking controller 3 on the vehicle drives to Park_Lot[1] according to the planned parking entry path.

[0069] User_B drives into the electronic fence, and the valet parking controller 3 on his vehicle shows a successful match. User_B continues to drive to the vicinity of Elevator[2].Park_Point[X][Y]. The valet parking controller 3 on his vehicle reminds the user to get off. After the user gets off, the valet parking controller 3 on his vehicle drives to Park_Lot[2] according to the planned parking entry path.

[0070] Similarly, the scenario where user User_A and user User_B drive out of the target parking lot.

[0071] When User_A is about to arrive near Gate[1], valet parking is initiated from the user's smart mobile terminal 1. The travel service provider 2 determines that Gate[1].Park_Point[X][Y] is not occupied and User_A has a higher ranking priority, and allocates Gate[1].Park_Point[X][Y] to User_A. The valet parking controller 3 on the vehicle controls the vehicle to drive from Park_Lot[1] to Gate[1].Park_Point[X][Y] according to the planned parking exit path.

[0072] When User_B is about to arrive near Elevator[2], valet parking is initiated from the user's smart mobile terminal 1. The travel service provider 2 determines that Elevator[2].Park_Point[X][Y] is not occupied and User_B has a higher ranking priority, and allocates Elevator[2].Park_Point[X][Y] to User_B. The valet parking controller 3 on the vehicle controls the vehicle to drive from Park_Lot[2] to Elevator[2].Park_Point[X][Y] according to the planned parking exit path.

[0073] like Figure 2 As shown in the figure, a valet parking service method is provided in an embodiment of the present invention, which is used in the valet parking service system of the embodiment of the present invention to realize automatic parking control of the vehicle. The method comprises the following steps:

[0074] Step S1, receiving one or more parking requests and post-parking requirements initiated by a vehicle;

[0075] Step S2: Calculate the priority data of each vehicle entering the corresponding target parking lot according to the parking request and the demand after parking initiated by each vehicle, and combine the first parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot to obtain the coordinates of the user's alighting point of each vehicle in the corresponding target parking lot and the corresponding allocated vacant parking space, and further plan the parking entry path of each vehicle from the coordinates of the user's alighting point to the allocated vacant parking space in the corresponding target parking lot;

[0076] Step S3: When it is detected that each vehicle has arrived at the coordinates of the user's getting-off point in the corresponding target parking lot and the user has gotten off the vehicle, each vehicle is controlled to automatically park in the corresponding allocated vacant parking space based on the parking entry path planned for each vehicle.

[0077] The specific process is as follows: in step S1, a parking request and post-parking demand initiated by one or more users for their vehicles are received; the post-parking demand is obtained in the following three ways: (1) the user actively inputs the post-parking demand in the valet parking APP, that is, the user directly inputs it; (2) after the user enters the destination in the valet parking APP, the APP automatically makes a demand judgment based on the user's destination type (restaurant, cinema, supermarket, etc.), that is, the demand is automatically judged based on the type of destination entered by the user; (3) after the user enters the parking lot near the destination in the valet parking APP, the APP actively recommends the destination and the user confirms it before making a demand judgment, that is, the parking lot near the destination entered by the user is actively recommended and the demand is judged after the user confirms it again.

[0078] In step S2, the time point of receiving the parking request and the parking demand (i.e., parking signal) initiated by the vehicle is used as the user sorting priority parameter A, the distance from the current position of the vehicle (i.e., the current driving position) to the corresponding target parking lot is used as the user sorting priority parameter B, and further the degree of fit between the user's destination and the resource information around the parking lot is used as the input parameter C, the mapping relationship between the resource information around the parking lot and the parking lot key coordinate point diagram information is used as the input parameter D, the time point when the user actually arrives near the parking point is used as the input parameter E, the mapping relationship between the current empty parking space and the parking lot key coordinate point diagram is used as the input parameter F, and the dynamic boarding / disembarking point array near the key coordinate point is used as the input parameter G. The above parameters are compiled into an array [A, B, C, D, E, F, G] uniquely associated with the user ID as the input of the artificial intelligence algorithm, the planned coordinates of the user's disembarking point in the parking lot {X1, Y1, Z1}, the allocated parking space number XXXX, and the parking entry path Path_Hope1[X][Y][Z] from the user's disembarking point to the allocated parking space.

[0079] In step S3, the planned path and parking lot key diagram information are downloaded. After the user is detected driving into the electronic fence, the key elements in the surrounding environment (navigation signs, parking space numbers, etc.) are identified through SLAM mapping and matched with the parking lot key diagram information. After the match is successful, the self-vehicle positioning is realized through wheel speed and IMU fusion positioning and the planned parking entry path starting point (i.e., the user's getting off point) is matched in real time. After the vehicle is detected to be about to drive to the user's getting off point, the user is prompted to get off. After the user is detected to get off, the vehicle control is started to enter the valet parking cruise stage. In the cruise stage, the valet parking controller continuously determines the current vehicle driving path Path_Now[X][Y][Z] through SLAM mapping matching, wheel speed and IMU fusion positioning, and performs real-time comparison with the planned parking entry path Path_Hope1[X][Y][Z]. If no insurmountable obstacle is detected by vision & radar fusion during the cruising phase, the vehicle is controlled to continue driving along the parking path Path_Hope1[X][Y][Z] until it parks in a pre-assigned vacant parking space. Conversely, if an insurmountable obstacle is detected by vision & radar fusion during the cruising phase, a request is made to replan the path and allocate a parking space, and the updated result is received after the request is received, and the vehicle is controlled again to perform automatic parking.

[0080] That is, when controlling each vehicle to perform automatic parking (driving in or out), the actual driving path of each vehicle is obtained through SLAM mapping matching and combined with the technology of wheel speed and IMU fusion positioning, and the actual driving path of each vehicle is compared in real time to see whether it is consistent with the corresponding planned parking entry path or parking exit path.

[0081] Through the fusion of vision and radar technology, it detects whether there are obstacles that cannot be bypassed on the actual driving path of each vehicle. When receiving early warning monitoring feedback from one or more vehicles that there are obstacles that cannot be bypassed on the actual driving path, all vehicles with early warning monitoring conditions are requested to re-plan the path, and further control all vehicles with early warning monitoring conditions to park in or out according to their respective re-planned paths.

[0082] It should be noted that the mapping relationships pre-built based on the key coordinate points of each target parking lot include: the mapping relationship between the key coordinate point diagram information of the parking lot and the resource information around the parking lot, the mapping relationship between the key coordinate point diagram information and the parking space location information, and the mapping relationship between the key coordinate points of the parking lot and the parking lot road boarding and alighting constraints. At this time, the coordinates of the user boarding and alighting points of each vehicle are based on the post-parking demand initiated by each vehicle, and the minimum pedestrian walking time and the minimum time for the vehicle to enter or exit the associated parking space are used as positioning conditions. The key coordinate points of each vehicle in the corresponding target parking lot are first located, and then determined by calculating the minimum pedestrian walking time from the located key coordinate points.

[0083] The priority data for each vehicle entering the corresponding target parking lot includes the time point when the parking request and the demand after parking are received for each vehicle, and the distance from the current position of each vehicle to the corresponding target parking lot; the priority data for each vehicle exiting the corresponding target parking lot includes the time point when the parking request and the demand after parking are received for each vehicle, and the distance from the current position of each vehicle to the expected boarding point of the user in the corresponding target parking lot.

[0084] In the embodiment of the present invention, the automatic parking and exit function of the vehicle can also be realized. Therefore, the method further includes:

[0085] The valet parking service system calculates the priority data of each vehicle exiting the corresponding target parking lot according to the parking request initiated for each vehicle and the demand after parking, and obtains the coordinates of the user boarding point of each vehicle in the corresponding target parking lot in combination with the second parameter combination data fed back by the mapping relationship pre-constructed based on the key coordinate points of each target parking lot, and further plans the parking exit path of each vehicle from the parking space position to the user boarding point coordinates in the corresponding target parking lot in combination with the parking space position of each vehicle in the corresponding target parking lot;

[0086] The valet parking service system controls each vehicle to automatically park and drive out to the corresponding user boarding point coordinates based on the parking exit path planned for each vehicle.

[0087] The specific process is as follows: first, the time point of receiving the parking request initiated by the vehicle and the demand after parking (i.e., parking signal) is used as the user sorting priority parameter A, the distance between the current parking space of the vehicle and the user's expected boarding point is used as the user sorting priority parameter B, the mapping relationship between the resource information around the parking lot and the simplified diagram information of the key coordinate points of the parking lot is used as the input parameter C, the mapping relationship between the simplified diagram information of the key coordinate points of the parking lot and the location information of the parking space is used as the input parameter D, and the array of dynamic boarding / disembarking points near the key coordinate points is used as the input parameter E. The above parameters are compiled into an array [A, B, C, D, E] uniquely associated with the user ID as the input of the artificial intelligence algorithm, and the output of the artificial intelligence algorithm is the parking exit path Path_Hope2[X][Y][Z] from the parking space to the user's boarding point {X2, Y2, Z2}.

[0088] Secondly, download the planned path and parking lot key diagram information. After downloading, start the vehicle to enter the cruising stage. Through SLAM mapping matching, wheel speed and IMU fusion positioning, the current vehicle position and parking exit path Path_Now[X][Y][Z] are continuously determined and compared with the planned parking exit path Path_Hope2[X][Y][Z] in real time. If no unavoidable obstacles are detected by visual & radar fusion in the cruising stage, the vehicle is controlled to continue to drive along the parking exit path Path_Hope2[X][Y][Z] until the coordinates {X2, Y2, Z2} pre-assigned to the user's boarding point are reached; conversely, if an unavoidable obstacle is detected by visual & radar fusion in the cruising stage, a request is made to re-plan the path, and the updated result is received after the request is received, and the vehicle is re-controlled to automatically park and exit.

[0089] Implementing the embodiments of the present invention has the following beneficial effects:

[0090] 1. The present invention adds a user's parking demand selection option based on the original collection of the user's intended parking location, collects the user's actual parking demand, and thus solves the problem of a single request data source in the existing valet parking method;

[0091] 2. In view of the problem that the allocation of free parking spaces is not in line with user expectations because only three factors, namely, the user queue priority, the distance between the free parking spaces and the driver's getting off point, and the presence or absence of obstacles between the free parking spaces and the driver's getting off point, are considered when allocating free parking spaces, this invention collects the first parameter combination data fed back by the mapping relationship pre-built by the target parking lot based on the key coordinate points (such as the mapping relationship between the resource information around the parking lot, the parking space location information, and the parking and getting off constraints on the parking lot road), and combines the collected user's parking needs to reasonably allocate free parking spaces through an artificial intelligence algorithm, which is more in line with the user's actual use needs, thereby solving the problem that the parking space allocation of the existing valet parking method does not meet the user's expectations, and can provide users with high-quality parking services;

[0092] 3. To address the parking congestion problem caused by fixed user alighting points, the present invention collects user parking needs, comprehensively considers the first parameter combination data fed back by the mapping relationship pre-built by the target parking lot based on key coordinate points (such as the mapping relationship between the resource information around the parking lot, the parking space location information, and the parking lot road alighting constraints), and obtains the dynamic user alighting point coordinates after comprehensive calculation, which greatly alleviates parking congestion.

[0093] It is worth noting that in the above system embodiment, the various system units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the various functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0094] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as ROM / RAM, disk, CD-ROM, etc.

[0095] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A valet parking service method, characterized in that: In a valet parking service system for realizing automatic parking of a vehicle, the method comprises the following steps: Receiving one or more parking requests and post-parking requirements initiated by a vehicle; According to the parking request initiated by each vehicle and the demand after parking, the priority data of each vehicle entering the corresponding target parking lot is calculated, and the coordinates of the user's alighting point of each vehicle in the corresponding target parking lot and the corresponding allocated vacant parking space are obtained in combination with the first parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot, and the parking entry path of each vehicle from the coordinates of the user's alighting point to the allocated vacant parking space in the corresponding target parking lot is further planned; When it is detected that each vehicle has arrived at the coordinates of the user's getting-off point in the corresponding target parking lot and the user has gotten off, based on the parking entry path planned for each vehicle, each vehicle is controlled to automatically park in the corresponding allocated vacant parking space; The mapping relationships pre-built based on the key coordinate points of each target parking lot include: the mapping relationship between the key coordinate point diagram information of the parking lot and the resource information around the parking lot, the mapping relationship between the key coordinate point diagram information of the parking lot and the parking space location information, and the mapping relationship between the key coordinate points of the parking lot and the parking lot road boarding and alighting constraints; In the mapping relationship between the simplified diagram information of the key coordinate points of the parking lot and the resource information around the parking lot, the priority is sorted according to the time it takes for a pedestrian to walk from the key node of the parking lot to the resource point in each mapping; In the mapping relationship between the parking lot key coordinate point diagram information and the parking space location information, the priority is sorted according to the time it takes for the vehicle to complete autonomous driving from the parking lot key node to the parking space under each mapping; In the mapping relationship between the key coordinate points of the parking lot and the parking lot road boarding and alighting constraints, priority is sorted according to the distance from the key nodes of the parking lot to the parking spaces under each mapping.

2. The valet parking service method according to claim 1, characterized in that: The method further comprises: According to the parking request initiated by each vehicle and the demand after parking, the priority data of each vehicle driving out of the corresponding target parking lot is calculated, and the coordinates of the user boarding point of each vehicle in the corresponding target parking lot are obtained by combining the second parameter combination data fed back by the mapping relationship pre-built based on the key coordinate points of each target parking lot, and further combined with the parking space position of each vehicle in the corresponding target parking lot, the parking exit path of each vehicle from the parking space position to the user boarding point coordinates in the corresponding target parking lot is planned; Based on the parking exit paths planned for each vehicle, each vehicle is controlled to automatically park and exit to the corresponding user boarding point coordinates.

3. The valet parking service method according to claim 2, characterized in that: The coordinates of the user alighting point and the user boarding point of each vehicle are based on the parking demand initiated by each vehicle, and the positioning conditions are the minimum walking time for pedestrians and the minimum time for vehicles to enter or exit the associated parking spaces. The key coordinate points of each vehicle in the corresponding target parking lot are first located, and then determined by calculating the minimum pedestrian walking time from the located key coordinate points.

4. The valet parking service method according to claim 2, characterized in that: The method further comprises: When controlling each vehicle to perform automatic parking, comparing in real time whether the actual driving path of each vehicle is consistent with the corresponding planned parking entry path or parking exit path; and, If one or more vehicles report that there are obstacles on the actual driving path that cannot be bypassed, a request is made to re-plan the routes for all vehicles with the warning monitoring conditions, and all vehicles with the warning monitoring conditions are further controlled to park in or out according to their respective re-planned routes.

5. The valet parking service method according to claim 4, characterized in that: The actual driving path of each vehicle is obtained through SLAM mapping matching combined with wheel speed and IMU fusion positioning technology; the early warning monitoring situation is detected through vision and radar fusion technology.

6. The valet parking service method according to claim 2, characterized in that: The priority data of each vehicle entering the corresponding target parking lot includes the time point of receiving the parking request initiated by each vehicle and the parking demand, and the distance from the current position of each vehicle to the corresponding target parking lot; The priority data for each vehicle to exit the corresponding target parking lot includes the time point at which the parking request and the parking demand initiated for each vehicle are received, and the distance from the current position of each vehicle to the user's expected boarding point in the corresponding target parking lot.

7. The valet parking service method according to claim 2, characterized in that: The parking demand is directly input by the user; or is automatically determined based on the type of destination input by the user; or is actively recommended based on parking lots near the destination input by the user and determined by the user again.

8. A valet parking service system, characterized in that: include: Smart mobile terminals, travel service providers and valet parking controllers; among them, There is at least one smart mobile terminal connected to the travel service provider; each smart mobile terminal is used to receive a parking request and a post-parking demand initiated by a user for the vehicle being driven and riding, and report the request to the travel service provider; The travel service provider includes a parking entry path planning unit; wherein the parking entry path planning unit is used to calculate the priority data of each vehicle entering the corresponding target parking lot according to the parking request initiated by each vehicle and the demand after parking, and obtain the user alighting point coordinates of each vehicle in the corresponding target parking lot and the corresponding allocated vacant parking space in combination with the first parameter combination data fed back by the mapping relationship pre-constructed based on the key coordinate points of each target parking lot, and plan the parking entry path of each vehicle from the user alighting point coordinates to the allocated vacant parking space in the corresponding target parking lot, and further send it to the valet parking controller; There is at least one valet parking controller, which is pre-installed on a vehicle driven by a corresponding user and then connected to the travel service provider; wherein each valet parking controller includes a parking entry control unit, which is used to control the vehicle to automatically park into a corresponding allocated vacant parking space based on a parking entry path planned for the vehicle when detecting that the vehicle has arrived at the coordinates of the user's getting-off point in the corresponding target parking lot and the user has gotten off the vehicle; The mapping relationships pre-built based on the key coordinate points of each target parking lot include: the mapping relationship between the key coordinate point diagram information of the parking lot and the resource information around the parking lot, the mapping relationship between the key coordinate point diagram information and the parking space location information, and the mapping relationship between the key coordinate points of the parking lot and the parking lot road boarding and alighting constraints; In the mapping relationship between the simplified diagram information of the key coordinate points of the parking lot and the resource information around the parking lot, the priority is sorted according to the time it takes for a pedestrian to walk from the key node of the parking lot to the resource point in each mapping; In the mapping relationship between the parking lot key coordinate point diagram information and the parking space location information, the priority is sorted according to the time it takes for the vehicle to complete autonomous driving from the parking lot key node to the parking space under each mapping; In the mapping relationship between the key coordinate points of the parking lot and the parking lot road boarding and alighting constraints, priority is sorted according to the distance from the key nodes of the parking lot to the parking spaces under each mapping.

9. The valet parking service system according to claim 8, characterized in that: The travel service provider further includes a parking exit path planning unit; wherein the parking exit path planning unit is used to calculate the priority data of each vehicle exiting the corresponding target parking lot according to the parking request initiated for each vehicle and the demand after parking, and to obtain the coordinates of the user boarding point of each vehicle in the corresponding target parking lot in combination with the second parameter combination data fed back by the mapping relationship pre-constructed based on the key coordinate points of each target parking lot, and to plan the parking exit path of each vehicle from the parking space position to the user boarding point coordinates in the corresponding target parking lot in combination with the parking space position of each vehicle in the corresponding target parking lot, and further send it to the valet parking controller; Each valet parking controller further includes a parking exit control unit, which is used to control the vehicle to automatically park and exit to the corresponding user boarding point coordinates based on the parking exit path corresponding to the vehicle.

Images