Valet parking and vehicle calling method and terminal equipment

Abstract

The invention discloses a valet parking and vehicle calling method and terminal equipment. The method comprises the following steps: acquiring a starting position and a destination position of a target vehicle; planning an accessible path set between the starting position and the destination position based on a parking lot geographic information model, wherein the parking lot geographic information model comprises entrance and exit information, lane connection relation information, lane trend information and parking space information; the comprehensive weight value of each path in the accessible path set is calculated, multiple accessible paths are selected from the accessible path set according to the comprehensive weight values, and the comprehensive weight values are used for representing the selectable degree of the paths; and traversing the plurality of accessible paths through a mileage counting output method, and determining a target path according to a traversing result. By implementing the technical scheme provided by the invention, the effect of improving the efficiency of valet parking and vehicle calling is achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation, and in particular to a valet parking and vehicle calling method and terminal equipment. Background Art

[0002] As modern transportation continues to move towards intelligence, valet parking and vehicle recall functions in the field of autonomous driving have become the focus of much attention. However, the current application status still has many defects that need to be solved.

[0003] In valet parking and vehicle calling scenarios, the complex parking environment poses many difficult problems for the path planning and execution of autonomous vehicles. The internal road layout of the parking lot often presents a non-standard layout, with many irregular changes and interruptions, and different areas also have traffic restrictions. This makes it difficult for vehicles to quickly and accurately plan an ideal driving path. Not only that, in scenarios where multiple vehicles are running at the same time, it is easy for them to gather at key sections or entrances and exits, forming a congested situation, which undoubtedly greatly prolongs the parking and recall time of the vehicle, seriously affecting the efficiency and experience of valet parking and vehicle calling services.

[0004] Therefore, how to improve the efficiency of valet parking and vehicle calling has become an urgent problem to be solved.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of this application is to provide a valet parking and vehicle calling method and terminal device, aiming to solve the technical problem of low efficiency of valet parking and vehicle calling.

[0007] To achieve the above-mentioned purpose, the present application provides a valet parking and vehicle calling method, which obtains the starting position and destination position of a target vehicle; plans a set of accessible paths between the starting position and the destination position based on a parking lot geographic information model, and the parking lot geographic information model includes entrance and exit information, lane connection relationship information, lane direction information, and parking space information; calculates a comprehensive weight value of each path in the accessible path set, and selects multiple accessible paths from the accessible path set according to the comprehensive weight value, and the comprehensive weight value is used to indicate the selectability of the path; traverses multiple accessible paths through a mileage counting output method, and determines a target path according to the traversal results.

[0008] Optionally, multiple accessible paths are traversed by a mileage counting output method, and a target path is determined based on the traversal results, including: segmenting the multiple accessible paths according to the areas through which the multiple accessible paths pass, to obtain a segmented path set; performing feature analysis on the segmented path set to determine the features of each segment in the segmented path set to obtain a segmented path feature set, wherein the feature analysis includes mileage analysis, curvature analysis, and traffic flow analysis; selecting multiple recommended sections from the segmented path feature set according to preset path selection criteria, and generating a target path based on the multiple recommended sections.

[0009] Optionally, before traversing multiple accessible paths through the mileage counting output method and determining the target path according to the traversal results, the method also includes: judging whether the number of multiple accessible paths is lower than the minimum number of accessible paths; if so, obtaining accessible features of the multiple accessible paths through a parking lot geographic information model, the accessible features including path length and passed area; merging the multiple accessible paths according to the accessible features to generate a new accessible path; merging the new accessible path into the multiple accessible paths, and updating the multiple accessible paths.

[0010] Optionally, after traversing multiple accessible paths through the mileage counting output method and determining the target path based on the traversal results, the method also includes: obtaining other target paths corresponding to other vehicles; comparing the target path with the other target paths to determine whether the target vehicle and the other vehicles pass through the same road section in the same time period; if so, adjusting the target path and the other target paths through a distributed collaborative algorithm, and sending the adjusted paths to the target vehicle and the other vehicles, respectively.

[0011] Optionally, before calculating the comprehensive weight value of each path in the accessible path set and selecting multiple accessible paths from the accessible path set based on the comprehensive weight values, the method also includes: predicting the number of non-autonomous driving vehicles on each path in the accessible path set through a behavior prediction model; and calculating the comprehensive weight value based on the number of non-autonomous driving vehicles.

[0012] Optionally, before predicting the number of non-autonomous driving vehicles on each path in the accessible path set through the behavior prediction model, the method also includes: obtaining historical driving data of non-autonomous driving vehicles, the historical driving data including driving data at different time periods, driving data at different dates, and driving data at different areas; training the initial neural network model through the historical driving data to obtain the behavior prediction model.

[0013] Optionally, the historical driving data includes driving trajectory, driving speed, and parking position. After obtaining the historical driving data of the non-autonomous driving vehicle, the method also includes: enhancing the historical driving data by a data enhancement method to obtain simulated driving data; updating the historical driving data by the simulated driving data, and training the initial neural network model by the updated historical driving data.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides a terminal device, which includes: an acquisition module, used to acquire the starting position and destination position of the target vehicle; a planning module, used to plan a set of accessible paths between the starting position and the destination position based on a parking lot geographic information model, the parking lot geographic information model including entrance and exit information, lane connection relationship information, lane direction information, and parking space information; a selection module, used to calculate the comprehensive weight value of each path in the accessible path set, and select multiple accessible paths from the accessible path set according to the comprehensive weight value, and the comprehensive weight value is used to indicate the degree of selectability of the path; a determination module, used to traverse the multiple accessible paths through a mileage counting output method, and determine the target path according to the traversal results.

[0015] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any method as in the first aspect.

[0016] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, the method steps of any one of the first aspects are executed.

[0017] The valet parking and vehicle calling method provided in the embodiment of the present application plans a set of accessible paths between a starting location and a destination location based on a geographic information model, calculates a comprehensive weight value for each path, and screens out multiple accessible paths with high accessibility based on the comprehensive weight value, thereby avoiding re-analyzing paths with low accessibility. Then, through a mileage counting output method, the road conditions of each section of the multiple accessible paths are analyzed to quickly determine a destination path with high accessibility, thereby improving the efficiency of valet parking and vehicle calling. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is an application flow chart provided in an embodiment of the present application.

[0019] Figure 2 It is a flow chart of the valet parking and vehicle calling method disclosed in the embodiment of the present application.

[0020] Figure 3It is a structural diagram of a terminal device disclosed in an embodiment of the present application.

[0021] Figure 4 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0022] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0025] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0026] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0027] As modern transportation continues to accelerate, the valet parking and vehicle recall functions in the field of autonomous driving have become the core focus of widespread attention both inside and outside the industry. With the rapid development of science and technology, people's expectations for transportation convenience and intelligence are rising, and the application of autonomous driving technology in many scenarios is highly anticipated, and valet parking and vehicle recall functions are regarded as key links to improve travel experience. However, in terms of current actual application status, its current application status is far from satisfactory, and there are many problems that need to be overcome urgently.

[0028] In valet parking and car-hailing scenarios, the internal roads of parking lots often present a non-standard layout, with many irregular changes and discontinuities, which poses many difficult problems for the path planning and execution of autonomous vehicles. For example, in order to make full use of space, some parking lots set up parking spaces of different shapes in different areas, which results in the vehicle driving passages having to meander and twist, with frequent corners and forks; and each area also has traffic restrictions, which must be taken into consideration when planning the route, which undoubtedly greatly increases the complexity and computational complexity of path planning, making it difficult for vehicles to quickly and accurately plan an ideal driving route.

[0029] Not only that, in scenarios where multiple vehicles are running at the same time, they tend to gather at key sections or entrances and exits, causing congestion. Especially in large and medium-sized parking lots, due to their large area and numerous parking spaces, there are often a large number of vehicles entering and exiting at the same time. During peak hours, vehicles may meet in narrow passages when driving towards parking spaces or exits, and autonomous vehicles may be stuck waiting for the other party to go first or give way. Entrances and exits, as the only way for vehicles to enter and exit, are the hardest hit by congestion. When vehicles are queuing to enter or leave the parking lot, once a vehicle breaks down or makes an operational error, it may trigger a chain reaction, paralyzing traffic at the entire entrance and exit. This congestion greatly prolongs the parking and recall time of vehicles, seriously affecting the efficiency and experience of valet parking and vehicle call-up services.

[0030] To this end, this application proposes a valet parking and vehicle calling method, which can be applied to valet parking and vehicle calling scenarios. Figure 1 , Figure 1 This is a flow chart provided by the embodiment of the present application. Figure 1In the process shown, when the owner needs the self-driving vehicle to park autonomously or to call a car, the owner sends the vehicle's starting position and destination position to the target parking lot. After receiving the starting position and destination position sent by the self-driving vehicle, the parking lot plans all accessible paths from the starting position to the destination position based on the parking lot's entrance and exit information, lane connection relationships, lane directions, and parking space coordinates, presets path selection criteria based on actual needs, selects multiple accessible paths from all accessible paths, and then traverses the multiple accessible paths through the mileage counting method to determine the target path, and sends the target path to the self-driving vehicle so that the vehicle can quickly drive to the destination position according to the target path.

[0031] based on Figure 1 Please continue to refer to the application scenarios shown in Figure 2 , Figure 2 This is a process of the valet parking and vehicle calling method provided in the embodiment of the present application. The valet parking and vehicle calling method can be applied to any terminal device that can run a program.

[0032] The following, combined Figure 2 A valet parking and vehicle calling method according to an embodiment of the present application is described in detail.

[0033] Step S201: Acquire the starting position and destination position of the target vehicle.

[0034] The starting position is used to indicate the initial position of the autonomous vehicle when performing valet parking and vehicle calling. In the valet parking scenario, the starting position can be the real-time position of the autonomous vehicle or any position entered by the vehicle owner.

[0035] The destination location is used to indicate the location that the autonomous vehicle should reach. The location can be a specific coordinate or an area. For example, in a valet parking scenario, the destination location can be area 1, which includes multiple parking spaces. The autonomous vehicle can park in any parking space in area 1.

[0036] Optionally, in a valet parking scenario, before obtaining the starting position and destination position of the target vehicle, the terminal device can send the available parking areas and vacant parking spaces near the parking lot to the autonomous driving vehicle, instructing the owner to select the initial position from the available parking area and the destination position from the vacant parking space.

[0037] Step S202: planning a set of accessible paths between the starting location and the destination location based on the parking lot geographic information model, where the parking lot geographic information model includes entrance and exit information, lane connection relationship information, lane direction information, and parking space information.

[0038] The parking lot geographic information model is used to store all-round information in the parking lot. In the embodiment of the present application, the parking lot geographic information model includes entrance and exit information, lane connection relationship information, lane direction information, and parking space information. Optionally, the parking lot geographic information model may also include obstacle distribution information, which is not specifically limited in the present application.

[0039] The reachable path set refers to all reachable paths from the starting location to the destination location.

[0040] Exemplarily, the entrance and exit information, lane connection relationship information, lane direction information and parking space information are obtained from the parking lot geographic information model, the starting position is defined as the starting node of the path search algorithm (such as the A* algorithm), and the destination position is set as the target node. The A* algorithm will gradually expand the search range according to the connection relationship between the nodes, the lane direction and the set heuristic function, from all possible paths from the starting node to the target node, and all possible paths are taken as the accessible path set. Optionally, each path consists of a series of continuous nodes, which correspond to key locations such as lanes, turning points, entrances and exits in the parking lot.

[0041] Step S203: Calculate the comprehensive weight value of each path in the accessible path set, and select multiple accessible paths from the accessible path set according to the comprehensive weight value. The comprehensive weight value is used to indicate the selectability of the path.

[0042] It is easy to understand that the higher the comprehensive weight value, the higher the cost of reaching the destination from the initial location. Therefore, when selecting multiple accessible paths from the accessible path set, all paths included in the accessible path set are sorted from low to high according to the comprehensive weight value, and the highest or lowest ranked accessible paths are selected according to the sorting results.

[0043] Exemplarily, the calculation factors for setting the comprehensive weight value include path length factor, curvature factor, and parking space availability factor, with weights of 40%, 30%, and 30% respectively. For example, for a path with a length of L, the number of bends is C, and the average curvature is K, the number of available parking spaces S within a certain range of the path (such as within 50 meters) is counted. Suppose the scores of a certain path in path length, curvature, and parking space availability factors are S1, S2, and S3 respectively, then S1 = (1-L / L_max) * 0.4, L_max is the length of the longest path in the accessible path set; S2 = (1-(C*K) / (C_max*K_max)) * 0.3, C_max and K_max are the maximum values ​​of the number of bends and the average curvature in the accessible path set respectively; S3 = S / S_max * 0.3, S_max is the maximum number of available parking spaces in the accessible path set; then its comprehensive weight value W = S1 + S2 + S3. The accessible path set is sorted from high to low according to the comprehensive weight value, and then multiple accessible paths with the highest ranking are selected according to the preset number of screening (for example, 5). These multiple accessible paths selected are relatively high selectivity paths selected after comprehensively considering factors such as path length, curvature, and parking space availability, and can therefore be used as candidate paths for the subsequent determination of the final target path.

[0044] It should be understood that, since the curvature values ​​of each path are different in different sections, the above average curvature is the average value of the sum of the curvature values ​​of each path in different sections.

[0045] Optionally, before calculating the comprehensive weight value of each path in the accessible path set and selecting multiple accessible paths from the accessible path set based on the comprehensive weight values, the method also includes: predicting the number of non-autonomous driving vehicles on each path in the accessible path set through a behavior prediction model; and calculating the comprehensive weight value based on the number of non-autonomous driving vehicles.

[0046] Exemplarily, after determining multiple passable paths, the information of the multiple passable paths is input into the behavior prediction model, and the behavior prediction model predicts the number of non-autonomous driving vehicles on each path based on the path information of the multiple passable paths; the weight ratio of the preset number of non-autonomous driving vehicles in the calculation of the comprehensive weight value is comprehensively calculated with other weight factors to obtain the comprehensive weight value. For example, the weight ratio of the number of non-autonomous driving vehicles is 25%, and for other traditional weight factors such as path length, curvature, and parking space availability, the weight ratios are 30%, 25%, and 20% respectively. A path scores 0.4 in the path length factor, 0.35 in the curvature factor, 0.28 in the parking space availability factor, and 0.53*0.25=0.1325 in the number of non-autonomous driving vehicles. The comprehensive weight value of the path is 0.4*0.3+0.35*0.25+0.28*0.2+0.1325=0.3055.

[0047] In this embodiment, the number of potential non-autonomous driving vehicles in an accessible path is predicted through a behavior prediction model, and the number of non-autonomous driving vehicles is included in the comprehensive weight calculation, which can effectively avoid paths where non-autonomous driving vehicles are concentrated, and reduce delays and risks caused by autonomous driving vehicles dealing with unpredictable human driving behavior.

[0048] Optionally, before predicting the number of non-autonomous driving vehicles on each path in the accessible path set through the behavior prediction model, the method also includes: obtaining historical driving data of non-autonomous driving vehicles, the historical driving data including driving data at different time periods, driving data at different dates, and driving data at different areas; training the initial neural network model through the historical driving data to obtain the behavior prediction model.

[0049] The initial neural network refers to a neural network model that has not been trained with data. This application uses the CNN-LSTM model as an example.

[0050] Exemplarily, the historical driving records of non-autonomous vehicles in the past year are extracted from the database of the parking lot management system. The historical driving records include the timestamps of the vehicle entering and leaving the parking lot, the vehicle's driving trajectory in the parking lot, the vehicle parking area, and the corresponding date information; the records are divided and stored according to different time periods (such as morning peak, noon, evening peak, night, etc.), different dates (weekdays, weekends, holidays, etc.), and different areas (areas near the parking lot entrance, internal channel areas, various parking partitions, etc.). The historical driving data is preprocessed, the time data is converted into digital codes, the regional data is regionally encoded, different parking areas and channel areas are given unique digital identifiers, and the vehicle trajectory data is normalized to improve the efficiency and stability of model training.

[0051] The architecture of the initial CNN-LSTM model includes two convolution layers. The first layer (CNN layer) has a convolution kernel size of 3x3, 64 kernels, and a step size of 1. It uses the ReLU activation function to extract spatial feature maps with different angles and certain nonlinear characteristics from historical driving data. The second layer has a convolution kernel size of 3x3, 128 kernels, and a step size of 1. It also uses the ReLU activation function to further combine and transform more types of features from the 64 feature maps output by the first layer to adapt to more complex changes in vehicle trajectory spatial patterns, capture the combination relationship between different trajectory segments, and the overall layout characteristics of the trajectory in a larger area. Each convolution layer is followed by a 2x2 maximum pooling layer to reduce the data dimension. Two LSTM layers are set up. The first LSTM layer receives the trajectory feature sequence processed by CNN and the time and area features as input, and performs preliminary learning and extraction of the time series features in these data, for example, capturing the time-related features such as the change in vehicle speed and the length of stay in different time periods; the second LSTM layer further deepens the learning of time series features based on the first layer, such as the long-term change law of vehicle driving behavior between different dates and different time periods. The various features processed by CNN and LSTM are integrated through the fully connected layer to form a new feature vector, and the ReLU activation function is continued to be used to continue to introduce nonlinear factors into the model to further enhance the model's learning ability for complex relationships. The output layer uses a linear activation function to process the feature vector integrated by the fully connected layer, and outputs a numerical value used to represent the predicted result of the number of non-autonomous driving vehicles on each accessible path.

[0052] The preprocessed historical driving data is divided into training set, validation set and test set. The ratio can be set to 70%, 20% and 10%. The initial CNN-LSTM model is trained, verified and tested through the training set, validation set and test set to obtain the behavior prediction model.

[0053] In this embodiment, the initial neural network is trained according to the characteristics of different time periods, different dates, different regions, etc. of the historical driving data to obtain a behavior prediction model, so that the behavior prediction model can predict the number of non-autonomous driving vehicles when querying the target path according to the time variation pattern of the number of non-autonomous driving vehicles, making the prediction result more accurate.

[0054] Optionally, the historical driving data includes driving trajectory, driving speed, and parking position. After obtaining the historical driving data of the non-autonomous driving vehicle, the method also includes: enhancing the historical driving data by a data enhancement method to obtain simulated driving data; updating the historical driving data by the simulated driving data, and training the initial neural network model by the updated historical driving data.

[0055] Take the data enhancement of driving trajectory data and parking position data as an example. For driving trajectory data, the random noise addition method is used; and for parking position data, the parking position offset method is used.

[0056] For example, at the coordinate point of the original driving trajectory, a random offset of 10% is added to each coordinate point, and the offset is randomly selected from a preset small range (such as [-0.5 meters, 0.5 meters]) in the horizontal and vertical directions. For example, the coordinates of the original trajectory point are (10.0, 20.0), which become (10.2, 19.9) after data enhancement; for each parking position coordinate, a new coordinate point is randomly selected from a small range around it (such as a circular area with a radius of 1 meter) as the simulated parking position. Through this data enhancement method, each historical driving data is processed to generate corresponding simulated driving data; then these simulated driving data are integrated with the original historical driving data to form a historical driving data containing original data and simulated data, and the initial neural network model is trained based on the historical driving data.

[0057] In this embodiment, by training the initial neural network model with updated historical driving data, the initial neural network model can learn more comprehensive and complex vehicle driving behavior characteristic patterns, enhance the accuracy and generalization ability of predicting the number of non-autonomous driving vehicles on accessible paths, and thus improve the reliability of the prediction results.

[0058] In an optional embodiment, before traversing multiple accessible paths through the mileage counting output method and determining the target path based on the traversal results, the method also includes: determining whether the number of multiple accessible paths is lower than the minimum number of accessible paths; if so, obtaining accessible features of the multiple accessible paths through a parking lot geographic information model, the accessible features including path length and passed area; merging the multiple accessible paths according to the accessible features to generate a new accessible path; merging the new accessible path into the multiple accessible paths, and updating the multiple accessible paths.

[0059] Exemplarily, assuming that the minimum number of accessible paths is set to 5, if the number of accessible paths currently counted is only 3, then it is determined that it is lower than the minimum number of accessible paths, and the path length of each path and the area information passed through, such as the entrance area, parking area A, channel connection area, construction area, etc., are obtained, and the area order and related attribute information are recorded, the overlapping parts and differences of the three paths are analyzed, and the three paths are connected and merged at reasonable nodes (such as the exit or entrance of a certain area that they pass through together) to form a new accessible path, and the newly generated accessible path is incorporated into the original set of multiple accessible paths to complete the update operation of the accessible path set.

[0060] In this embodiment, the number of paths in the accessible path set is increased, and the new path combines the advantages of multiple previous paths, improving the quality and diversity of the overall accessible paths, and providing a richer and higher-quality selection basis for subsequent traversal to determine a better target path through the mileage counting output method.

[0061] Step S204: traverse multiple accessible paths through the mileage counting output method, and determine the target path according to the traversal results.

[0062] Exemplarily, an independent mileage counting variable is created for each of a plurality of accessible paths (e.g., 5), and detailed information of the 5 accessible paths is obtained, including lane information of each segment of the path, turning point information, entrance and exit information, etc. The path with the highest comprehensive weight value among the 5 accessible paths (the first path) is used as the path to be traversed currently, starting from the starting position of the first path, all segmented paths contained in the path are determined, and the path length, turning points, special road conditions (such as slope, speed limit changes, etc.) and other information of each segmented path are obtained, and the preset variable values ​​corresponding to these information are accumulated, and the accumulated result is used as the segmented path value of the segmented path, and then the segmented path values ​​of all the segmented paths are summed up as the traversal value of the first path; based on the traversal method for the first path, the other 4 accessible paths are continued to be traversed, the traversal values ​​of the other 4 accessible paths are calculated, the traversal values ​​of the 5 accessible paths are compared, and the path with the lowest traversal value is used as the destination path.

[0063] In the above example, the segmented path is obtained by dividing multiple accessible paths according to the regional layout of the parking lot (such as different parking areas, passage areas, entrance and exit areas, etc.). For example, the path is divided according to the main regional nodes passed through. If a path goes from the entrance of the parking lot through the central passage to a certain parking area, then the entrance to the central passage entrance section, the central passage section, and the central passage exit to the parking area section can be used as different segments. Optionally, the path can also be divided into multiple lane sections according to a preset length, which is not specifically limited in this application.

[0064] The valet parking and vehicle calling method provided in the embodiment of the present application plans a set of accessible paths between a starting location and a destination location based on a geographic information model, calculates a comprehensive weight value for each path, and screens out multiple accessible paths with high accessibility based on the comprehensive weight value, thereby avoiding re-analyzing paths with low accessibility. Then, through a mileage counting output method, the road conditions of each section of the multiple accessible paths are analyzed to quickly determine a destination path with high accessibility, thereby improving the efficiency of valet parking and vehicle calling.

[0065] The above example traverses multiple accessible paths based on the mileage counting output method, and selects one path from the multiple accessible paths as the target path.

[0066] In another optional example, multiple accessible paths are traversed by a mileage counting output method, and a target path is determined based on the traversal results, including: segmenting the multiple accessible paths according to the areas through which the multiple accessible paths pass, to obtain a segmented path set; performing feature analysis on the segmented path set to determine the features of each segment in the segmented path set to obtain a segmented path feature set, wherein the feature analysis includes mileage analysis, curvature analysis, and traffic flow analysis; selecting multiple recommended sections from the segmented path feature set according to preset path selection criteria, and generating a target path based on the multiple recommended sections.

[0067] The preset path selection criteria are the criteria set before selecting the recommended route. For example, the preset path selection criteria may be a criterion for selecting the recommended route based on a comprehensive consideration of multiple factors such as route length, curvature, and traffic flow, with corresponding weights set for each factor, and a weighted calculation of the comprehensive route score to screen the recommended route based on the score and generate the target route.

[0068] As described above, multiple accessible paths are segmented according to the regional layout of the parking lot, a unique identifier is established for each segmented path, and its sequence information in the original accessible path is stored to form a segmented path set.

[0069] Exemplarily, the mileage of the segmented path is calculated by querying the lane length data in the parking lot geographic information model. For example, if a segmented path contains three consecutive lanes with lengths of 50 meters, 30 meters and 40 meters respectively, then the mileage of the segmented path is 120 meters; the curvature of each segment of the path is determined using the lane direction information and geometric calculation methods. For example, the curvature of a curve with a radius of 20 meters is calculated to be 1 / 20=0.05; the real-time traffic monitoring system of the parking lot or the traffic prediction model based on historical data is interacted with to obtain the traffic flow information of the area where each segment of the path is located. For example, a segmented path is located in the main passage of the parking lot and is during peak hours. According to historical data, the traffic volume in this area is predicted to be 100 vehicles per hour, then the traffic flow value of the segmented path is recorded as 100. The weight of mileage is set to 40%, the weight of curvature is set to 30%, and the weight of traffic flow is set to 30%. The comprehensive score of each path is calculated based on these weights. For example, the mileage score S1, curvature score S2, and traffic flow score S3 of a segmented path, the comprehensive score W = S1*0.4+S2*0.3+S3*0.3. Then, the segmented path feature set is sorted according to the comprehensive score, and multiple segmented paths with higher scores are selected as recommended segments. According to the order information of the recommended segments in the original accessible path and the connection relationship between them, these recommended segments are combined to generate the target path.

[0070] In this example, by segmenting multiple accessible paths and combining them to generate a target path based on feature analysis, the advantageous sections of each accessible path can be integrated, thereby reducing the driving time of the vehicle in the parking lot and improving the efficiency of valet parking and vehicle calling.

[0071] In an optional embodiment, after traversing multiple accessible paths through the mileage counting output method and determining the target path based on the traversal results, the method also includes: obtaining other target paths corresponding to other vehicles; comparing the target path with the other target paths to determine whether the target vehicle and the other vehicles pass through the same road section in the same time period; if so, adjusting the target path and the other target paths through a distributed collaborative algorithm, and sending the adjusted paths to the target vehicle and the other vehicles, respectively.

[0072] Other target paths are used to represent the paths of other autonomous vehicles from the starting location to the destination location.

[0073] Exemplarily, after determining the destination path corresponding to the target vehicle, obtain other target path information corresponding to other vehicles, compare and analyze the target path of the target vehicle with other target paths, and for each lane in the target vehicle path, check whether the target path of other vehicles also passes through the section within the same 5-minute time window. For example, the target vehicle is scheduled to pass through a certain section of the central passage of the parking lot from 10:00 to 10:05. If the target path of other vehicles shows that it also passes through the section within this time period, it is determined that there is a conflict. After determining that there is a conflicting path, the distributed collaborative algorithm selects one of the vehicles (assuming that the target vehicle is selected) according to the set priority order (such as giving priority to adjusting the vehicle path with the longest path), finds other channels or areas around it, plans a new path segment, inserts it into the original target path, and forms the adjusted target path.

[0074] In this implementation, the conflicting target paths and other target paths are adjusted through a distributed collaborative algorithm, which can effectively avoid interference caused by vehicles independently planning their paths, optimize and integrate parking lot traffic flow, reduce the situation where vehicles are forced to stop or frequently adjust their routes due to path conflicts, and improve overall traffic efficiency and stability.

[0075] It is understandable that the terminal device is required to achieve Figure 2The functions described include hardware and / or software modules that perform the corresponding functions. In combination with the steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware 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 in combination with the embodiments, but such implementation should not be considered to be beyond the scope of the present application.

[0076] In this embodiment, the terminal device can be divided into functional modules according to the above method example. For example, different functional modules can be divided according to different functions, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0077] In the case of dividing each functional module into corresponding functional modules, Figure 3 A possible schematic diagram of the terminal device 300 involved in the above embodiment is shown, and the terminal device 300 includes: an acquisition module 301, used to acquire the starting position and the destination position of the target vehicle; a planning module 302, used to plan a set of accessible paths between the starting position and the destination position based on a parking lot geographic information model, the parking lot geographic information model including entrance and exit information, lane connection relationship information, lane direction information, and parking space information; a selection module 303, used to calculate the comprehensive weight value of each path in the accessible path set, and select multiple accessible paths from the accessible path set according to the comprehensive weight value, and the comprehensive weight value is used to indicate the selectability of the path; a determination module 304, used to traverse the multiple accessible paths through a mileage counting output method, and determine the target path according to the traversal results.

[0078] In an optional implementation of the embodiment of the present application, the determination module 304 is also used to segment the multiple accessible paths according to the areas through which the multiple accessible paths pass, to obtain a segmented path set; perform feature analysis on the segmented path set to determine the characteristics of each segment of the path in the segmented path set to obtain a segmented path feature set, the feature analysis including mileage analysis, curvature analysis, and traffic flow analysis; select multiple recommended sections from the segmented path feature set according to preset path selection criteria, and generate a target path based on the multiple recommended sections.

[0079] In an optional implementation of the embodiment of the present application, the determination module 304 is also used to determine whether the number of multiple accessible paths is lower than the minimum number of accessible paths; if so, obtain the accessible characteristics of the multiple accessible paths through the parking lot geographic information model, the accessible characteristics including path length and passed area; merge the multiple accessible paths according to the accessible characteristics to generate a new accessible path; merge the new accessible path into the multiple accessible paths, and update the multiple accessible paths.

[0080] In an optional implementation method of the embodiment of the present application, the determination module 304 is also used to obtain other target paths corresponding to other vehicles; compare the target path and other target paths to determine whether the target vehicle and other vehicles pass through the same road section in the same time period; if so, adjust the target path and other target paths through a distributed collaborative algorithm, and send the adjusted paths to the target vehicle and other vehicles respectively.

[0081] In an optional implementation of the embodiment of the present application, the selection module 303 is also used to predict the number of non-autonomous driving vehicles on each path in the accessible path set through a behavior prediction model; and calculate a comprehensive weight value based on the number of non-autonomous driving vehicles.

[0082] In an optional implementation of the embodiment of the present application, the selection module 303 is also used to obtain historical driving data of non-autonomous driving vehicles, and the historical driving data includes driving data of different time periods, driving data of different dates, and driving data of different areas; the initial neural network model is trained through the historical driving data to obtain a behavior prediction model.

[0083] In an optional implementation of the embodiment of the present application, the selection module 303 is also used to enhance the historical driving data through a data enhancement method to obtain simulated driving data; update the historical driving data through the simulated driving data, and train the initial neural network model through the updated historical driving data.

[0084] The present application also discloses an electronic device. Figure 4 , Figure 4 The electronic device 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .

[0085] The communication bus 402 is used to realize the connection and communication between these components.

[0086] The user interface 403 may include a display screen (Display) and a camera (Camera), and the optional user interface 403 may also include a standard wired interface and a wireless interface.

[0087] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0088] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or more combinations of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented by a single chip.

[0089] Among them, the memory 405 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may also be optionally at least one storage device located away from the aforementioned processor 401. Refer to Figure 4 , the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program for a valet parking and vehicle calling method.

[0090] exist Figure 4 In the electronic device 400 shown, the user interface 403 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 401 can be used to call the application program storing the valet parking and vehicle calling method in the memory 405. When executed by one or more processors 401, the electronic device 400 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0091] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0092] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0093] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0094] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0095] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0096] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A valet parking and vehicle calling method, characterized in that: include: Get the starting position and destination position of the target vehicle; Planning a set of accessible paths between the starting location and the destination location based on a parking lot geographic information model, wherein the parking lot geographic information model includes entrance and exit information, lane connection relationship information, lane direction information, and parking space information; Calculating a comprehensive weight value of each path in the accessible path set, and selecting a plurality of accessible paths from the accessible path set according to the comprehensive weight value, wherein the comprehensive weight value is used to indicate the selectability of the path; The plurality of accessible paths are traversed by a mileage counting output method, and a target path is determined according to the traversal results.

2. The method according to claim 1, characterized in that The traversing the plurality of accessible paths by the mileage counting output method and determining the target path according to the traversal results includes: Segmenting the plurality of accessible paths according to the areas through which the plurality of accessible paths pass, to obtain a segmented path set; Performing feature analysis on the segmented path set to determine the features of each segment of the path in the segmented path set to obtain a segmented path feature set, wherein the feature analysis includes mileage analysis, curvature analysis, and traffic flow analysis; A plurality of recommended sections are selected from the segmented path feature set according to a preset path selection criterion, and the target path is generated based on the plurality of recommended sections.

3. The method according to claim 1, characterized in that Before traversing the plurality of accessible paths by the mileage counting output method and determining the target path according to the traversal results, the method further includes: Determining whether the number of the plurality of accessible paths is less than a minimum number of accessible paths; If it is lower than that, obtaining the passable features of the plurality of accessible paths through the parking lot geographic information model, wherein the passable features include the path length and the passing area; According to the passable characteristics, the multiple accessible paths are merged to generate a new accessible path; The new accessible path is merged into the plurality of accessible paths, and the plurality of accessible paths is updated.

4. The method according to claim 1, characterized in that: After traversing the plurality of accessible paths by the mileage counting output method and determining the target path according to the traversal results, the method further includes: Get other target paths corresponding to other vehicles; Comparing the target path with the other target paths, determining whether the target vehicle and the other vehicles pass through the same road section in the same time period; If so, the target path and the other target paths are adjusted by a distributed collaborative algorithm, and the adjusted paths are sent to the target vehicle and the other vehicles respectively.

5. The method according to claim 1, characterized in that Before calculating the comprehensive weight value of each path in the accessible path set and selecting a plurality of accessible paths from the accessible path set according to the comprehensive weight value, the method further includes: Predicting the number of non-autonomous driving vehicles on each path in the accessible path set by a behavior prediction model; The comprehensive weight value is calculated based on the number of the non-autonomous driving vehicles.

6. The method according to claim 5, characterized in that Before predicting the number of non-autonomous driving vehicles on each path in the accessible path set by using the behavior prediction model, the method further includes: Acquire historical driving data of the non-autonomous driving vehicle, wherein the historical driving data includes driving data of different time periods, driving data of different dates, and driving data of different regions; The initial neural network model is trained using the historical driving data to obtain the behavior prediction model.

7. The method according to claim 6, characterized in that The historical driving data includes a driving trajectory, a driving speed, and a parking position. After acquiring the historical driving data of the non-automatic driving vehicle, the method further includes: The historical driving data is enhanced by a data enhancement method to obtain simulated driving data; The historical driving data is updated by the simulated driving data, and the initial neural network model is trained by the updated historical driving data.

8. A terminal device, characterized in that: The terminal device comprises: An acquisition module is used to acquire the starting position and destination position of the target vehicle; A planning module, for planning a set of accessible paths between the starting location and the destination location based on a parking lot geographic information model, wherein the parking lot geographic information model includes entrance and exit information, lane connection relationship information, lane direction information, and parking space information; A selection module, configured to calculate a comprehensive weight value of each path in the accessible path set, and select a plurality of accessible paths from the accessible path set according to the comprehensive weight value, wherein the comprehensive weight value is used to indicate the selectability of the path; The determination module is used to traverse the multiple accessible paths through a mileage counting output method, and determine a target path according to the traversal results.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images