Route recommendation method, route recommendation model training method and device

Abstract

The present disclosure provides a path recommendation method, a training method and device for a path recommendation model, data processing technology and map technology, specifically relating to autonomous driving, intelligent transportation and deep learning, which can be applied to scenarios such as vehicle navigation and path planning. The specific implementation scheme is: obtaining a vehicle driving path recommendation request, determining a recommendation domain sequence based on the vehicle driving path recommendation request, wherein the recommendation domain sequence is a recommended driving transfer relationship between vehicles between domains, and the domain includes a named domain determined based on a named road segment with the same road name in the road network, and a virtual domain determined based on an unnamed road segment in the road network, generating and outputting the vehicle driving path based on the recommendation domain sequence, that is, determining the vehicle driving path with "domain" as the granularity. Compared with training with "road segment" as the granularity, the "solution space" of path recommendation is relatively small, which can achieve the technical effect of improving recommendation efficiency.

Description

Technical Field

[0001] The present disclosure relates to data processing technology and map technology, specifically to autonomous driving, intelligent transportation, and deep learning, and can be applied to scenarios such as vehicle navigation and path planning. In particular, it relates to a path recommendation method, a training method for a path recommendation model, and a device. Background Art

[0002] With the intelligentization of vehicle travel, vehicles can travel from their starting point to their destination through navigation.

[0003] In related technologies, historical vehicle driving paths may be acquired to recommend vehicle driving paths based on the acquired vehicle driving paths. Summary of the Invention

[0004] The present disclosure provides a path recommendation method, a training method and a device for a path recommendation model for improving recommendation efficiency.

[0005] According to a first aspect of the present disclosure, a route recommendation method is provided, comprising:

[0006] Get vehicle driving route recommendation request;

[0007] Determining a recommended domain sequence based on the vehicle driving route recommendation request, wherein the recommended domain sequence is a recommended driving transfer relationship between the vehicle domains, and the domains include named domains determined based on named road segments with the same road name in a road network, and virtual domains determined based on unnamed road segments in the road network;

[0008] A vehicle driving path is generated and output according to the recommended domain sequence.

[0009] According to a second aspect of the present disclosure, a method for training a route recommendation model is provided, comprising:

[0010] Aggregating named road segments with the same road name in a road network to obtain a named domain; aggregating unnamed road segments in the road network to obtain a virtual domain;

[0011] Determining a vehicle travel transfer relationship between domains based on the road network and the obtained sample vehicle travel paths traveling on the road network, wherein the domains include the named domains and the virtual domains;

[0012] A route recommendation model is obtained by training according to the driving transfer relationship and the sample vehicle driving path, wherein the route recommendation model is used to recommend a vehicle driving path.

[0013] According to a third aspect of the present disclosure, a route recommendation device is provided, comprising:

[0014] A first acquiring unit is configured to acquire a vehicle driving route recommendation request;

[0015] a first determining unit, configured to determine a recommended domain sequence based on the vehicle driving route recommendation request, wherein the recommended domain sequence is a recommended driving transfer relationship between domains for the vehicle, and the domains include named domains determined based on named road segments with the same road name in a road network, and virtual domains determined based on unnamed road segments in the road network;

[0016] a generating unit, configured to generate a vehicle driving path according to the recommended domain sequence;

[0017] An output unit is used to output the vehicle driving path.

[0018] According to a fourth aspect of the present disclosure, a training device for a route recommendation model is provided, comprising:

[0019] an aggregation unit for aggregating named road segments with the same road name in the road network to obtain a named domain; and aggregating unnamed road segments in the road network to obtain a virtual domain;

[0020] a second determining unit, configured to determine a travel transfer relationship between vehicles between domains based on the road network and the obtained sample vehicle travel paths traveling on the road network, wherein the domains include the named domains and the virtual domains;

[0021] A training unit is used to train a path recommendation model based on the driving transfer relationship and the sample vehicle driving path, wherein the path recommendation model is used to recommend a vehicle driving path.

[0022] According to a fifth aspect of the present disclosure, there is provided an electronic device, including:

[0023] at least one processor; and

[0024] a memory communicatively connected to the at least one processor; wherein,

[0025] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect or the second aspect.

[0026] According to a sixth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method according to the first aspect or the second aspect.

[0027] According to the seventh aspect of the present disclosure, a computer program product is provided, comprising: a computer program, wherein the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program so that the electronic device executes the method described in the first aspect or the second aspect.

[0028] According to the present disclosure, a recommendation domain sequence representing the recommended driving transfer relationship is determined based on the named domain and the virtual domain, and a vehicle driving path is determined based on the recommendation domain sequence, that is, the technical feature of determining the vehicle driving path with the "domain" as the granularity is relatively small compared to training with the "road section" as the granularity, which can achieve the technical effect of improving the recommendation efficiency.

[0029] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0031] Figure 1 is a schematic diagram according to a first embodiment of the present disclosure;

[0032] Figure 2 is a schematic diagram according to a second embodiment of the present disclosure;

[0033] Figure 3 is a schematic diagram according to a third embodiment of the present disclosure;

[0034] Figure 4 is a schematic diagram according to a fourth embodiment of the present disclosure;

[0035] Figure 5 is a schematic diagram according to a fifth embodiment of the present disclosure;

[0036] Figure 6 is a schematic diagram according to a sixth embodiment of the present disclosure;

[0037] Figure 7 is a schematic diagram according to a seventh embodiment of the present disclosure;

[0038] Figure 8 is a schematic diagram according to an eighth embodiment of the present disclosure;

[0039] Figure 9 is a schematic diagram according to a ninth embodiment of the present disclosure;

[0040] Figure 10It is a block diagram of an electronic device used to implement the route recommendation method and the training method of the route recommendation model according to the embodiment of the present disclosure. DETAILED DESCRIPTION

[0041] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0042] As vehicle travel becomes more intelligent, vehicle users have higher demands for vehicle travel safety and efficiency. To meet vehicle users' vehicle travel needs and improve vehicle travel safety and efficiency, at least the following embodiments can be used to recommend vehicle travel routes.

[0043] In some embodiments, a vehicle driving path with the minimum total cost may be searched and recommended based on a graph theory algorithm.

[0044] Exemplarily, the travel cost of each road in the road network is mined based on historical driving trajectories or road attributes, and a graph theory algorithm is used to accelerate the search for a path with the minimum travel cost, and the path with the minimum travel cost is determined as the recommended vehicle driving path.

[0045] Road attributes include road type, such as expressways, and road congestion levels. The cost of a route can be set by the user based on their needs. For example, the lowest cost can be the shortest travel time, the smoothest route, or the lowest toll.

[0046] However, the travel cost of a vehicle's driving path is not completely equivalent to the cumulative sum of the costs of the roads and intersections along the vehicle's driving path. The road sequence relationship in the vehicle's driving path is also a factor affecting the cost. For example, the travel cost of merging from a secondary road into the main road and then turning left is different from the travel cost of turning left on the main road. Therefore, the above method has the disadvantage of low accuracy in recommending vehicle driving paths.

[0047] In other embodiments, a route recommendation model may be trained to generate and recommend a vehicle driving route based on the route recommendation model.

[0048] Exemplarily, sample data is obtained, where the sample data is a historical vehicle driving path, a basic network model is trained based on the sample data to obtain a path recommendation model, and a vehicle driving path is recommended based on the path recommendation model so that the vehicle can travel based on the vehicle driving path, or so that the user can control the vehicle driving based on the vehicle driving path.

[0049] However, the road network covers a wide area, including millions or even more road segments. Both training the path recommendation model and predicting the vehicle driving path based on the path recommendation model have the disadvantages of low efficiency and high resource consumption.

[0050] In order to avoid the above technical problems, the inventors of the present disclosure have obtained the inventive concept of the present disclosure through creative work: determining a recommended domain sequence according to a vehicle driving path recommendation request, that is, determining the recommended driving transfer relationship between the vehicle domains, so as to generate a vehicle driving path based on the transfer relationship.

[0051] Based on the above-mentioned inventive concept, the present disclosure provides a path recommendation method, a training method and a device for a path recommendation model, which involve data processing technology and map technology, specifically autonomous driving, intelligent transportation and deep learning, and can be applied to scenarios such as vehicle navigation and path planning to improve the efficiency and reliability of vehicle driving path recommendations.

[0052] Figure 1 is a schematic diagram according to the first embodiment of the present disclosure, as shown in Figure 1 As shown, the path recommendation method of the embodiment of the present disclosure includes:

[0053] S101: Obtain a vehicle driving route recommendation request.

[0054] Exemplarily, the execution subject of this embodiment can be a path recommendation device (hereinafter referred to as the recommendation device), and the recommendation device can be a server (such as a cloud server, or a local server, or a server cluster, etc.), or a computer, or a terminal device, or a processor, or a chip, etc., which will not be listed one by one here.

[0055] This embodiment does not limit the method of obtaining the vehicle driving route recommendation request. For example, it can be obtained based on a vehicle-mounted terminal set up with the vehicle, or based on a user device of a vehicle user, or based on a sound pickup device set up in the vehicle, etc., which are not listed one by one here.

[0056] S102: Determine a recommended domain sequence based on the vehicle driving route recommendation request. The recommended domain sequence is a recommended driving transfer relationship between domains for the vehicle. The domains include named domains determined based on named road segments with the same road name in the road network, and virtual domains determined based on unnamed road segments in the road network.

[0057] A road network can be understood as a road system consisting of various road segments that are interconnected and interwoven into a network. For example, a road network includes road segments, which may or may not have road names, and different road segments may have the same road name.

[0058] A named domain can be understood as an area consisting of named road segments with the same road name, while a virtual domain can be understood as an area consisting of unnamed road segments.

[0059] The recommended driving transfer relationship and the sample driving transfer relationship described later are relative concepts. “Recommended” and “sample” cannot be understood as limitations on the driving transfer relationship.

[0060] For example, if a vehicle travels from one domain to another, the travel relationship between the two domains can be called a travel transfer relationship. For example, if a vehicle can travel from domain A to domain B, then a travel transfer relationship can be called a travel transfer relationship between domains A and B.

[0061] And since the domain includes a named domain and a virtual domain, the corresponding travel transfer relationship can be a travel transfer relationship between a named domain and a named domain, a travel transfer relationship between a named domain and a virtual domain, or a travel transfer relationship between a virtual domain and a virtual domain.

[0062] S103: Generate and output a vehicle driving path according to the recommended domain sequence.

[0063] Based on the above analysis, it can be seen that the embodiment of the present disclosure provides a path recommendation method, including: obtaining a vehicle driving path recommendation request, determining a recommendation domain sequence according to the vehicle driving path recommendation request, wherein the recommendation domain sequence is a recommended driving transfer relationship between vehicles between domains, and the domain includes a named domain determined based on a named road segment with the same road name in the road network, and a virtual domain determined based on an unnamed road segment in the road network, generating and outputting a vehicle driving path according to the recommendation domain sequence. In this embodiment, a recommendation domain sequence representing the recommended driving transfer relationship is determined based on the named domain and the virtual domain, and the vehicle driving path is determined based on the recommendation domain sequence, that is, the technical feature of determining the vehicle driving path with "domain" as the granularity. Compared with training with "road segment" as the granularity, the "solution space" of path recommendation is relatively small, which can achieve the technical effect of improving recommendation efficiency.

[0064] In order to make readers more deeply understand the implementation principle of this disclosure, Figure 2 For example Figure 1 The illustrated embodiment is explained in more detail.

[0065] Figure 2 is a schematic diagram according to the second embodiment of the present disclosure, as shown in Figure 2 As shown, the path recommendation method of the embodiment of the present disclosure includes:

[0066] S201: Obtain a vehicle driving route recommendation request, where the driving route recommendation request includes a driving start point and a driving end point.

[0067] It should be understood that in order to avoid cumbersome descriptions, the technical features of this embodiment that are the same as those in the above embodiments will not be described in detail in this embodiment.

[0068] The starting point is the starting point of the vehicle's travel, and the end point is the destination of the vehicle's travel.

[0069] S202: Determine a recommended domain sequence based on the travel start point and travel end point. The recommended domain sequence is a recommended travel transition relationship between domains for the vehicle. The domains include named domains determined based on named road segments with the same road name in the road network, and virtual domains determined based on unnamed road segments in the road network.

[0070] For example, the domain (including named domain and / or unnamed domain) between the driving starting point and the driving end point can be determined, and the driving transfer relationship between the domains as the vehicle travels from the driving starting point to the driving end point can be determined. This driving transfer relationship can be called a recommended driving transfer relationship.

[0071] In some embodiments, the named domain is obtained by aggregating named road segments with the same road name in the road network; and the virtual domain is obtained by aggregating unnamed road segments in the road network.

[0072] For example, the recommendation device can determine named road segments and unnamed road segments from the road network. For each named road segment, the recommendation device can obtain the road name corresponding to each named road segment, extract named road segments with the same road name from each named road segment, and aggregate the extracted named road segments to obtain a named domain. The named domain can be understood as an area consisting of named road segments with the same road name.

[0073] For example, if the names of the named road segments A1, A2, and A3 are all A, then the named road segments A1, A2, and A3 can be aggregated to obtain a named domain. Accordingly, the named domain can be understood as a connected area composed of the named road segments A1, A2, and A3.

[0074] Similarly, for each unnamed road segment, the recommendation device can aggregate the unnamed road segments to obtain a virtual domain. The virtual domain can be understood as an area composed of unnamed road segments.

[0075] In this embodiment, by combining aggregation processing to obtain named domains and unnamed domains, the determined named domains and unnamed domains can have higher accuracy and reliability, thereby achieving the technical effect of higher accuracy and reliability of the vehicle driving path recommended with "domain" as the granularity.

[0076] In some embodiments, there are multiple unnamed road sections; the virtual domain is determined based on the obtained undirected graph corresponding to the road network, with any unnamed road section as the starting point of exploration, until a named road section is explored from the undirected graph, and the virtual domain represents the area between any unnamed road section and the explored named road section.

[0077] In other words, the unnamed road segments can be aggregated according to the connected components of the undirected graph to create a virtual domain.

[0078] For example, in the undirected graph corresponding to the road network, the surrounding area is explored starting from the current unnamed road segment until a named road segment is explored, and a virtual domain is created including the area between the starting point (i.e., the current unnamed road segment) and the explored named road segment, and so on until all virtual domains are obtained.

[0079] In this embodiment, the virtual domain is obtained by combining undirected graph exploration, so that the virtual domain can have higher accuracy and effectiveness.

[0080] S203: Generate and output a vehicle driving path based on the recommended domain sequence and a preset mapping relationship, wherein the mapping relationship is used to represent a mapping relationship between named road sections and road names in a road network.

[0081] In some embodiments, the mapping relationship is obtained by obtaining the road names corresponding to the respective named road sections from the road network, and constructing the corresponding relationship between the named road sections and the road names.

[0082] Exemplarily, the recommended domain sequence is the recommended driving transfer relationship between vehicle domains. The domains include named domains and unnamed domains, and the named domains are obtained by aggregating named road sections with the same road names. Therefore, after the recommended domain sequence is determined, the vehicle driving path can be determined from the granularity of the "domain" in combination with the mapping relationship, thereby improving the efficiency of determining the vehicle driving path.

[0083] In some embodiments, S203 may include: determining a road segment corresponding to each domain in the recommended domain sequence according to the mapping relationship, and generating and outputting a vehicle driving path according to each determined road segment.

[0084] By combining the mapping relationship, the recommended domain sequence at the "domain" granularity can be converted into the vehicle driving path at the "road section" granularity, so as to recommend a relatively more "fine-grained" vehicle driving path, meet the user's vehicle driving needs, and improve the safety and reliability of vehicle driving.

[0085] In some embodiments, the vehicle driving route recommendation request also includes user characteristics of the vehicle user. The user characteristics are used to characterize the age, gender, and driving preferences of the vehicle user, so as to determine the vehicle driving route in combination with the user characteristics, thereby meeting the personalized needs of the vehicle user and improving the diversity and flexibility of the recommendation.

[0086] In some embodiments, a path recommendation model can also be constructed to recommend a vehicle driving path based on the path recommendation model. For example, a vehicle driving path recommendation request (such as including a driving starting point and a driving end point, and also including user features) can be input into the path recommendation model, and a recommendation domain sequence can be output to generate and output the vehicle driving path based on the recommendation domain sequence and the mapping relationship.

[0087] Exemplarily, the driving start point, driving end point and user features are spliced ​​to obtain spliced ​​features, and the spliced ​​features are input into the path recommendation model to output a recommended domain sequence to generate and output the vehicle driving path based on the recommended domain sequence and the mapping relationship.

[0088] Now combined Figure 3 The training method of the path recommendation model is described, Figure 3 is a schematic diagram according to the third embodiment of the present disclosure, as shown in Figure 3 As shown, the training method of the path recommendation model of the embodiment of the present disclosure includes:

[0089] S301: Aggregate named road segments with the same road name in the road network to obtain a named domain. Aggregate unnamed road segments in the road network to obtain a virtual domain.

[0090] Illustratively, the execution subject of this embodiment is a training device for the path recommendation model (hereinafter referred to as the training device). The training device can be the same device as the path recommendation device or a different device, which is not limited in this embodiment.

[0091] Similarly, in order to avoid cumbersome descriptions, the technical features of this embodiment that are the same as those in the above embodiments will not be repeated in this embodiment.

[0092] In some embodiments, the training device can determine named road segments and unnamed road segments from the road network. For each named road segment, the training device can obtain the road name corresponding to each named road segment, extract named road segments with the same road name from each named road segment, and aggregate the extracted named road segments to obtain a named domain. The named domain can be understood as an area consisting of named road segments with the same road name.

[0093] Similarly, for each unnamed road section, the training device can aggregate the unnamed road sections to obtain a virtual domain. The virtual domain can be understood as an area composed of unnamed road sections.

[0094] S302: Determine the driving transfer relationship between the vehicles in the domains based on the road network and the obtained sample vehicle driving paths on the road network, wherein the domains include named domains and virtual domains.

[0095] The sample vehicle driving paths can be understood as vehicle driving paths generated by vehicles driving on the road network. This embodiment does not limit the number of sample vehicle driving paths, and can be determined based on demand, historical records, and experiments.

[0096] For example, for scenarios with relatively high training requirements, the number of sample vehicle driving paths may be relatively large, and for scenarios with relatively low training requirements, the number of sample vehicle driving paths may be relatively small.

[0097] This embodiment can be understood as follows: Based on the road system and the sample vehicle travel path, it can be determined that a vehicle travels from one domain to another. The travel relationship between a vehicle traveling from one domain to another can be referred to as a travel transfer relationship between one domain and another. For example, if a vehicle can travel from domain A to domain B, a travel transfer relationship can be referred to as a travel transfer relationship between domains A and B.

[0098] And since the domain includes a named domain and a virtual domain, the corresponding travel transfer relationship can be a travel transfer relationship between a named domain and a named domain, a travel transfer relationship between a named domain and a virtual domain, or a travel transfer relationship between a virtual domain and a virtual domain.

[0099] S303: A route recommendation model is obtained based on the driving transfer relationship and the sample vehicle driving routes. The route recommendation model is used to recommend vehicle driving routes.

[0100] Based on the above analysis, it can be seen that the embodiment of the present disclosure provides a training method for a path recommendation model, including: aggregating named road sections with the same road name in a road network to obtain named domains. Aggregating unnamed road sections in a road network to obtain virtual domains, determining the driving transfer relationship of vehicles between domains based on the road network and the obtained sample vehicle driving paths, wherein the domain includes a named domain and a virtual domain, and training a path recommendation model based on the driving transfer relationship and the sample vehicle driving paths. The path recommendation model is used to recommend vehicle driving paths. In this embodiment, the technical feature of the path recommendation model is obtained by generating named domains and virtual domains to determine the driving transfer relationship of vehicles between domains, that is, training with "domain" as the granularity. Compared with training with "road section" as the granularity, the "solution space" of training is relatively small, which can improve the technical effect of training efficiency.

[0101] In order to make readers more deeply understand the implementation principle of this disclosure, Figure 4 The above embodiments are described in more detail.

[0102] in, Figure 4 is a schematic diagram according to a fourth embodiment of the present disclosure, as shown in Figure 4 As shown, the training method of the path recommendation model of the embodiment of the present disclosure includes:

[0103] S401: Obtain the road name corresponding to each road segment from the road network, and aggregate the named road segments with the same road name to obtain a named domain.

[0104] It should be understood that in order to avoid cumbersome descriptions, the technical features of this embodiment that are the same as those in the above embodiments will not be described in detail in this embodiment.

[0105] S402: Obtain an undirected graph corresponding to the road network, and obtain unnamed road sections without road names in the road network.

[0106] An undirected graph includes points and edges. A point can be understood as an intersection in a road network, and an edge between two points can be understood as a road segment between the two points.

[0107] S403: Taking any unnamed road segment as the starting point of exploration, until a named road segment is explored from the undirected graph, and determining the area between any unnamed road segment and the explored named road segment as a virtual domain.

[0108] In other words, the unnamed road segments can be aggregated according to the connected components of the undirected graph to create a virtual domain.

[0109] For example, in the undirected graph corresponding to the road network, the surrounding area is explored starting from the current unnamed road segment until a named road segment is explored, and a virtual domain is created including the area between the starting point (i.e., the current unnamed road segment) and the explored named road segment, and so on until all virtual domains are obtained.

[0110] In this embodiment, the virtual domain is obtained by combining undirected graph exploration, so that the virtual domain can have higher accuracy and effectiveness.

[0111] S404: Determine the driving transfer relationship between vehicles in the domains based on the road network and the obtained sample vehicle driving paths on the road network, and generate a new undirected graph based on the driving transfer relationship.

[0112] The domain includes the named domain and the virtual domain.

[0113] Exemplarily, this step can be understood as updating the undirected graph involved in S402. In S402, the granularity of the wireless graph is a road segment, while the granularity of the new undirected graph is a domain.

[0114] In some embodiments, determining the driving transfer relationship between vehicles between domains based on the road network and the sample vehicle driving paths obtained on the road network can include: extracting the connection relationship between the domains in the road network, and determining the driving transfer relationship based on the connection relationship and the sample vehicle driving paths.

[0115] Exemplarily, the driving transfer relationship is determined based on the connection relationship between named domains, the connection relationship between named domains and unnamed domains, and the connection relationship between unnamed domains, so that the driving transfer relationship is the transfer information of vehicle driving with domain as the granularity.

[0116] Since the connection relationship can be between named domains, between unnamed domains, or between a named domain and an unnamed domain, the driving transfer relationship determined by combining the connection relationship has a technical effect of high accuracy and reliability.

[0117] In some embodiments, a path recommendation model can be trained based on a new undirected graph, driving transfer relations, and sample vehicle driving paths, that is, a path recommendation model can be trained with "domain" as the granularity to improve training efficiency.

[0118] S405: Constructing a mapping relationship between named road sections and road names.

[0119] For example, combined with the above analysis, one road name may correspond to multiple named road sections. Therefore, in the mapping relationship, one road name may correspond to one named road section or multiple named road sections. However, usually one named road section corresponds to one road name.

[0120] S406: Based on the new undirected graph, the driving transition relationship, the mapping relationship, and the sample vehicle driving paths, a route recommendation model is trained to recommend vehicle driving paths.

[0121] Combined with the above analysis, the path recommendation model obtained through new undirected graph training can improve training efficiency. Since the mapping relationship can represent the correspondence between named road sections and road names, the path recommendation model obtained by combining mapping relationship training can achieve relatively fine-grained training, thereby improving the accuracy and effectiveness of the trained path recommendation model.

[0122] In some embodiments, S406 may include the following steps:

[0123] The first step is to determine the trajectory sequence of each road section in the road network based on the sample vehicle driving path.

[0124] The road segment trajectory sequence is used to represent the road segments that the vehicle travels on and the sequential relationship between the road segments.

[0125] The second step: According to the mapping relationship and the new undirected graph, the road segment trajectory sequence is converted into a domain trajectory sequence, and the path recommendation model is trained based on the road segment trajectory sequence, the driving transfer relationship and the domain trajectory sequence.

[0126] Among them, the road segment trajectory sequence and the domain trajectory sequence are relative concepts. The road segment trajectory sequence is a trajectory sequence with the road segment as the granularity, and the domain trajectory sequence is a trajectory sequence with the domain as the granularity. Since the domain is obtained by aggregating the road segments, such as the named domain is obtained by aggregating the named road segments with the same road name, and the virtual domain is obtained by aggregating the unnamed road segments, therefore, relatively speaking, the training based on the domain trajectory sequence converges faster and is more efficient.

[0127] In some embodiments, converting the segment trajectory sequence into the domain trajectory sequence according to the mapping relationship and the new undirected graph may include the following steps:

[0128] The first step: for each road segment in the road segment trajectory sequence, determine the domain corresponding to each road segment according to the mapping relationship.

[0129] Exemplarily, a road segment trajectory sequence may include named road segments and unnamed road segments. The mapping relationship can characterize the correspondence between named road segments and road names, and the named domain is obtained based on the aggregation of named road segments with the same road names. Therefore, for the named road segments in the road segment trajectory sequence, the corresponding named domain can be determined according to the mapping relationship, and for the unnamed road segments in the road segment trajectory sequence, the unnamed domain corresponding to the unnamed road segments can be determined.

[0130] The second step is to obtain the connectivity relationship between the domains corresponding to each road segment according to the new undirected graph, and generate a domain trajectory sequence according to the connectivity relationship.

[0131] Combined with the above analysis, the new undirected graph is an undirected graph with domain as the granularity. By combining the new undirected graph to generate domain trajectory sequences, the accuracy and reliability of the domain trajectory sequences can be improved.

[0132] In some embodiments, the road segment trajectory sequence has a trajectory starting point and a trajectory end point; training a path recommendation model based on the road segment trajectory sequence, the driving transfer relationship, and the domain trajectory sequence may include: training a path recommendation model based on the trajectory starting point, the trajectory end point, the driving transfer relationship, and the domain trajectory sequence.

[0133] Since different trajectory starting points and trajectory end points may correspond to different vehicle driving paths, in order to improve the accuracy and reliability of vehicle driving path recommendation, a path recommendation model can be obtained by combining the trajectory starting points and trajectory end points for training.

[0134] In some embodiments, user characteristics of vehicle users corresponding to the road segment trajectory sequence can also be obtained, that is, user characteristics of vehicle users forming the sample vehicle driving path. User characteristics are used to characterize the user's age, gender, and driving preferences, etc., and a path recommendation model is obtained by combining user feature training to meet the personalized needs of path recommendation and improve the flexibility and diversity of path recommendation.

[0135] For example, user features, trajectory starting points, trajectory ending points, and domain trajectory sequences may be subjected to feature splicing processing to obtain splicing features, and a route recommendation model may be trained based on the driving transfer relationship and the splicing features.

[0136] Among them, the splicing features can be used as input features, and the input features can be predicted based on the driving transfer relationship to obtain the predicted vehicle driving path. By calculating the loss value between the predicted vehicle driving path and the pre-calibrated real vehicle driving path, iterative training is performed based on the loss value to obtain a path recommendation model.

[0137] Figure 5 is a schematic diagram according to a fifth embodiment of the present disclosure, as shown in Figure 5 As shown, the path recommendation device 500 of the embodiment of the present disclosure includes:

[0138] The first acquiring unit 501 is configured to acquire a vehicle driving route recommendation request.

[0139] The first determining unit 502 is configured to determine a recommended domain sequence based on a vehicle driving route recommendation request, wherein the recommended domain sequence is a recommended driving transfer relationship between the vehicle domains, and the domains include named domains determined based on named road segments with the same road name in the road network, and virtual domains determined based on unnamed road segments in the road network.

[0140] The generating unit 503 is configured to generate a vehicle driving path according to the recommended domain sequence.

[0141] The output unit 504 is configured to output the vehicle's driving path.

[0142] Figure 6 is a schematic diagram according to a sixth embodiment of the present disclosure, as shown in Figure 6 As shown, the path recommendation device 600 of the embodiment of the present disclosure includes:

[0143] The first acquiring unit 601 is configured to acquire a vehicle driving route recommendation request.

[0144] In some embodiments, the vehicle driving route recommendation request includes a driving starting point, a driving end point, and user characteristics of the vehicle user making the vehicle driving route recommendation request.

[0145] The first determination unit 602 is configured to determine a recommended domain sequence based on a vehicle driving route recommendation request, wherein the recommended domain sequence is a recommended driving transfer relationship between the vehicle domains, and the domains include named domains determined based on named road segments with the same road name in the road network, and virtual domains determined based on unnamed road segments in the road network.

[0146] Combine Figure 6 It can be seen that, in some embodiments, the first determining unit 602 includes:

[0147] The input subunit 6021 is used to input the vehicle driving route recommendation request into the pre-trained route recommendation model.

[0148] The output subunit 6022 is used to output the recommended domain sequence.

[0149] Among them, the path recommendation model is trained based on the sample driving transfer relationship and the sample vehicle driving path. The sample driving transfer relationship is the driving transfer relationship between vehicles between domains determined based on the road network and the obtained sample vehicle driving path.

[0150] The generating unit 603 is configured to generate a vehicle driving path according to the recommended domain sequence.

[0151] In some embodiments, the generating unit 603 is used to generate a vehicle driving path according to the recommended domain sequence and a preset mapping relationship, wherein the mapping relationship is used to represent a mapping relationship between a named road section and a road name in a road network.

[0152] Combine Figure 6 It can be seen that, in some embodiments, the generating unit 603 includes:

[0153] The first determining subunit 6031 is configured to determine the road segment corresponding to each domain in the recommended domain sequence according to the mapping relationship.

[0154] The first generating subunit 6032 is configured to generate a vehicle driving path according to the determined road sections.

[0155] In some embodiments, the mapping relationship is obtained by obtaining the road names corresponding to the respective named road sections from the road network, and constructing the corresponding relationship between the named road sections and the road names.

[0156] In some embodiments, the named domain is obtained by aggregating named road segments with the same road name in the road network; and the virtual domain is obtained by aggregating unnamed road segments in the road network.

[0157] In some embodiments, there are multiple unnamed road sections; the virtual domain is determined based on the obtained undirected graph corresponding to the road network, with any unnamed road section as the starting point of exploration, until a named road section is explored from the undirected graph, and the virtual domain represents the area between any unnamed road section and the explored named road section.

[0158] The output unit 604 is used to output the vehicle driving path.

[0159] Figure 7 7 is a schematic diagram according to a seventh embodiment of the present disclosure. As shown in the figure, a training device 700 for a path recommendation model according to an embodiment of the present disclosure includes:

[0160] The aggregation unit 701 is used to aggregate named road segments with the same road name in the road network to obtain a named domain; and to aggregate unnamed road segments in the road network to obtain a virtual domain.

[0161] The second determining unit 702 is configured to determine a vehicle travel transfer relationship between domains based on the road network and the obtained sample vehicle travel paths traveling on the road network, wherein the domains include named domains and virtual domains.

[0162] The training unit 703 is used to train a route recommendation model based on the driving transfer relationship and the sample vehicle driving routes, wherein the route recommendation model is used to recommend vehicle driving routes.

[0163] Figure 8 is a schematic diagram according to an eighth embodiment of the present disclosure, as shown in Figure 8 As shown, the training device 800 of the path recommendation model according to the embodiment of the present disclosure includes:

[0164] The aggregation unit 801 is used to aggregate named road segments with the same road name in the road network to obtain a named domain; and to aggregate unnamed road segments in the road network to obtain a virtual domain.

[0165] In some embodiments, there are multiple unnamed road segments; the aggregation unit 801 includes:

[0166] The acquisition subunit 8011 is used to obtain an undirected graph corresponding to the road network.

[0167] The exploration subunit 8012 is used to use any unnamed road segment as the exploration starting point until a named road segment is explored from the undirected graph.

[0168] The third determining subunit 8013 is configured to determine an area between any unnamed road segment and the discovered named road segment as a virtual domain.

[0169] The second determining unit 802 is configured to determine a vehicle travel transfer relationship between domains based on the road network and the obtained sample vehicle travel paths traveling on the road network, wherein the domains include named domains and virtual domains.

[0170] Combine Figure 8 It can be seen that, in some embodiments, the second determining unit 802 includes:

[0171] The extraction subunit 8021 is used to extract the connection relationship between each domain in the road network.

[0172] The second determining subunit 8022 is configured to determine a driving transfer relationship based on the connection relationship and the sample vehicle driving path.

[0173] The second acquiring unit 803 is configured to acquire the road name corresponding to each named road section from the road network.

[0174] The construction unit 804 is used to construct a mapping relationship between named road sections and road names.

[0175] The third acquiring unit 805 is configured to acquire user characteristics of the vehicle user corresponding to the road segment trajectory sequence.

[0176] The training unit 806 is used to train a route recommendation model based on the driving transfer relationship and the sample vehicle driving routes, wherein the route recommendation model is used to recommend the vehicle driving route.

[0177] Combine Figure 8 It can be seen that, in some embodiments, the training unit 806 includes:

[0178] The second generating subunit 8061 is used to generate an undirected graph corresponding to the road network according to the driving transfer relationship.

[0179] The training subunit 8062 is used to train a route recommendation model based on the undirected graph, the driving transfer relationship, and the sample vehicle driving routes.

[0180] The training subunit 8062 is used to train a route recommendation model based on the undirected graph, the driving transfer relationship, the mapping relationship, and the sample vehicle driving path.

[0181] In some embodiments, the training subunit 8062 includes:

[0182] The determination module is used to determine the segment trajectory sequence of the vehicle traveling on each road segment in the road network based on the sample vehicle driving path.

[0183] The conversion module is used to convert the road segment trajectory sequence into the domain trajectory sequence according to the undirected graph and mapping relationship.

[0184] In some embodiments, the conversion module includes:

[0185] The determination submodule is used to determine the domain corresponding to each road section in the road section trajectory sequence according to the mapping relationship.

[0186] The acquisition subunit is used to obtain the connectivity relationship between the domains corresponding to the determined road sections according to the undirected graph.

[0187] The generation submodule is used to generate domain trajectory sequences based on connectivity relations.

[0188] The training module is used to train a path recommendation model based on the road segment trajectory sequence, driving transfer relationship and domain trajectory sequence.

[0189] In some embodiments, the road segment trajectory sequence has a trajectory start point and a trajectory end point; the training module is used to train a path recommendation model based on the trajectory start point, the trajectory end point, the driving transfer relationship and the domain trajectory sequence.

[0190] The training module is used to train a route recommendation model based on user characteristics, trajectory starting point, trajectory end point, driving transfer relationship and domain trajectory sequence.

[0191] In some embodiments, the training module includes:

[0192] The splicing submodule is used to perform feature splicing processing on user features, trajectory starting points, trajectory ending points, and domain trajectory sequences to obtain splicing features.

[0193] The training submodule is used to train a route recommendation model based on driving transfer relations and splicing features.

[0194] Figure 9 is a schematic diagram according to the ninth embodiment of the present disclosure, as shown in Figure 9 As shown, the electronic device 900 in the present disclosure may include: a processor 901 and a memory 902 .

[0195] Memory 902 is used to store programs. Memory 902 may include volatile memory (volatile memory), such as random-access memory (RAM), such as static random-access memory (SRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), etc. Memory may also include non-volatile memory (non-volatile memory), such as flash memory. Memory 902 is used to store computer programs (such as applications and functional modules that implement the above-mentioned methods), computer instructions, etc. The above-mentioned computer programs and computer instructions may be partitioned and stored in one or more memories 902. Furthermore, the above-mentioned computer programs, computer instructions, data, etc. may be called by processor 901.

[0196] The aforementioned computer programs, computer instructions, etc. may be partitioned and stored in one or more memories 902 . Furthermore, the aforementioned computer programs, computer instructions, etc. may be called by the processor 901 .

[0197] The processor 901 is configured to execute the computer program stored in the memory 902 to implement the various steps in the method involved in the above embodiment.

[0198] For details, please refer to the relevant description in the previous method embodiment.

[0199] The processor 901 and the memory 902 may be independent structures or integrated structures. When the processor 901 and the memory 902 are independent structures, the memory 902 and the processor 901 may be coupled via a bus 903 .

[0200] The electronic device of this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principles are the same and will not be repeated here.

[0201] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information (such as user characteristics, etc.) involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0202] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0203] According to an embodiment of the present disclosure, the present disclosure also provides a computer program product, which includes: a computer program, the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device executes the solution provided by any of the above embodiments.

[0204] Figure 10 A schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0205] like Figure 10 As shown, the device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the device 1000 can also be stored in the RAM 1003. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0206] Various components in device 1000 are connected to I / O interface 1005, including an input unit 1006, such as a keyboard, mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a magnetic disk, optical disk, etc.; and a communication unit 1009, such as a network card, modem, wireless communication transceiver, etc. The communication unit 1009 allows device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0207] The computing unit 1001 can be a variety of general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 1001 performs the various methods and processes described above, such as the route recommendation method and the training method of the route recommendation model. For example, in some embodiments, the route recommendation method and the training method of the route recommendation model can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as the storage unit 1008. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the route recommendation method and the training method of the route recommendation model described above can be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute the path recommendation method or the path recommendation model training method in any other appropriate manner (eg, by means of firmware).

[0208] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0209] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0210] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0211] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0212] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0213] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact via a communication network. This client-server relationship is established by computer programs running on the respective computers, establishing a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host, a host product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosts and VPS services ("Virtual Private Servers" or simply "VPS"). The server may also be a server in a distributed system or a server integrated with blockchain.

[0214] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0215] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A route recommendation method, comprising: Get vehicle driving route recommendation request; Determining a recommended domain sequence based on the vehicle driving route recommendation request, wherein the recommended domain sequence is a recommended driving transfer relationship between the vehicle domains, and the domains include named domains determined based on named road segments with the same road name in a road network, and virtual domains determined based on unnamed road segments in the road network; Generate and output a vehicle driving path according to the recommended domain sequence; The named domain is obtained by aggregating named road segments with the same road name in the road network; the virtual domain is obtained by aggregating unnamed road segments in the road network; Among them, there are multiple unnamed road sections; the virtual domain is determined based on the obtained undirected graph corresponding to the road network, and with any unnamed road section as the exploration starting point until a named road section is explored from the undirected graph, and the virtual domain represents the area between any unnamed road section and the explored named road section.

2. The method according to claim 1, wherein The generating and outputting the vehicle driving path according to the recommended domain sequence includes: The vehicle driving path is generated and output according to the recommended domain sequence and a preset mapping relationship, wherein the mapping relationship is used to represent the mapping relationship between famous road sections and road names in a road network.

3. The method according to claim 2, wherein: The generating and outputting the vehicle driving path according to the recommended domain sequence and the preset mapping relationship includes: The road sections corresponding to the respective domains in the recommended domain sequence are determined according to the mapping relationship, and the vehicle driving path is generated and output according to the determined road sections.

4. The method according to claim 2 or 3, wherein the mapping relationship is obtained by obtaining the road names corresponding to each named road section from the road network, and constructing the corresponding relationship between the named road sections and the road names.

5. The method according to any one of claims 1 to 3, wherein: The determining of a recommended domain sequence according to the vehicle driving path recommendation request includes: The vehicle driving route recommendation request is input into a pre-trained route recommendation model, and the recommended domain sequence is output, wherein the route recommendation model is trained based on sample driving transfer relations and sample vehicle driving routes obtained on the road network, and the sample driving transfer relations are driving transfer relations between vehicles in the domains determined based on the road network and the sample vehicle driving routes.

6. The method according to claim 5, wherein: The vehicle driving route recommendation request includes a driving starting point, a driving end point, and user characteristics of the vehicle user making the vehicle driving route recommendation request.

7. The method according to claim 5, wherein: The training method of the path recommendation model is: Aggregating named road segments with the same road name in the road network to obtain a named domain; aggregating unnamed road segments in the road network to obtain a virtual domain; Determining a vehicle travel transfer relationship between domains based on the road network and the obtained sample vehicle travel paths traveling on the road network, wherein the domains include the named domains and the virtual domains; A route recommendation model is obtained by training according to the driving transfer relationship and the sample vehicle driving path, wherein the route recommendation model is used to recommend a vehicle driving path.

8. The method according to claim 7, wherein: The route recommendation model is obtained by training according to the driving transfer relationship and the sample vehicle driving path, including: generating an undirected graph corresponding to the road network according to the driving transfer relationship; The route recommendation model is obtained by training according to the undirected graph, the driving transfer relationship and the sample vehicle driving path.

9. The method according to claim 8, further comprising: Obtaining the road names corresponding to the respective named road sections from the road network, and establishing a mapping relationship between the named road sections and the road names; Furthermore, the path recommendation model is obtained by training according to the undirected graph, the driving transfer relationship and the sample vehicle driving path, including: the path recommendation model is obtained by training according to the undirected graph, the driving transfer relationship, the mapping relationship and the sample vehicle driving path.

10. The method according to claim 9, wherein: The training of the route recommendation model according to the undirected graph, the driving transfer relationship, the mapping relationship, and the sample vehicle driving path includes: Determining a sequence of segment trajectories of the vehicle traveling on each segment in the road network based on the sample vehicle travel paths; According to the undirected graph and the mapping relationship, the road segment trajectory sequence is converted into a domain trajectory sequence, and the path recommendation model is trained based on the road segment trajectory sequence, the driving transfer relationship and the domain trajectory sequence.

11. The method according to claim 10, wherein: The converting the road segment trajectory sequence into a domain trajectory sequence according to the undirected graph and the mapping relationship includes: For each road segment in the road segment trajectory sequence, determining a domain corresponding to each road segment according to the mapping relationship; The connectivity relationship between the domains corresponding to the determined road sections is obtained according to the undirected graph, and the domain trajectory sequence is generated according to the connectivity relationship.

12. The method according to claim 10 or 11, wherein: The segment trajectory sequence has a trajectory start point and a trajectory end point; The training of the route recommendation model according to the road segment trajectory sequence, the driving transfer relationship, and the domain trajectory sequence includes: The path recommendation model is trained based on the trajectory starting point, the trajectory end point, the driving transfer relationship, and the domain trajectory sequence.

13. The method according to claim 12, further comprising: Obtaining user characteristics of vehicle users corresponding to the road segment trajectory sequence; And, the path recommendation model is trained based on the trajectory starting point, the trajectory end point, the driving transfer relationship and the domain trajectory sequence, including: the path recommendation model is trained based on the user characteristics, the trajectory starting point, the trajectory end point, the driving transfer relationship and the domain trajectory sequence.

14. The method according to claim 13, wherein: The training of the route recommendation model according to the user characteristics, the trajectory starting point, the trajectory ending point, the driving transfer relationship, and the domain trajectory sequence includes: Performing feature splicing processing on the user feature, the trajectory starting point, the trajectory ending point, and the domain trajectory sequence to obtain a splicing feature; The route recommendation model is obtained by training according to the driving transfer relationship and the splicing features.

15. The method according to any one of claims 7 to 11, 13 to 14, wherein: The determining, based on the road network and the obtained sample vehicle driving paths traveling on the road network, the driving transfer relationship between the vehicles in the domains includes: The connection relationship between each domain in the road network is extracted, and the driving transfer relationship is determined according to the connection relationship and the driving path of the sample vehicle.

16. The method according to any one of claims 7 to 11, 13 to 14, wherein: There are multiple unnamed road sections; the unnamed road sections in the road network are aggregated to obtain a virtual domain, including: Obtaining an undirected graph corresponding to the road network; Taking any unnamed road segment as the starting point of exploration, the process continues until a named road segment is explored from the undirected graph, and determining the area between the any unnamed road segment and the explored named road segment as a virtual domain.

17. A route recommendation device, comprising: A first acquiring unit is configured to acquire a vehicle driving route recommendation request; a first determining unit, configured to determine a recommended domain sequence based on the vehicle driving route recommendation request, wherein the recommended domain sequence is a recommended driving transfer relationship between domains for the vehicle, and the domains include named domains determined based on named road segments with the same road name in a road network, and virtual domains determined based on unnamed road segments in the road network; a generating unit, configured to generate a vehicle driving path according to the recommended domain sequence; An output unit, configured to output the vehicle's driving path; The named domain is obtained by aggregating named road segments with the same road name in the road network; the virtual domain is obtained by aggregating unnamed road segments in the road network; There are multiple unnamed road sections; the virtual domain is determined based on an undirected graph obtained corresponding to the road network, with any unnamed road section as the exploration starting point, until a named road section is explored from the undirected graph, and the virtual domain represents the area between any unnamed road section and the explored named road section.

18. The device according to claim 17, wherein The generating unit is configured to generate the vehicle driving path according to the recommended domain sequence and a preset mapping relationship, wherein the mapping relationship is used to represent a mapping relationship between named road sections and road names in a road network.

19. The device according to claim 18, wherein The generating unit includes: a first determining subunit, configured to determine, according to the mapping relationship, a road segment corresponding to each domain in the recommended domain sequence; The first generating subunit is used to generate the vehicle driving path according to the determined road sections.

20. The device according to claim 18 or 19, wherein the mapping relationship is obtained by obtaining the road names corresponding to the respective named road sections from the road network, and constructing the corresponding relationship between the named road sections and the road names.

21. The device according to any one of claims 17-18, wherein The first determining unit includes: An input subunit, configured to input the vehicle driving route recommendation request into a pre-trained route recommendation model; an output subunit, configured to output the recommended domain sequence; Among them, the path recommendation model is trained based on the sample driving transfer relationship and the sample vehicle driving path obtained on the road network. The sample driving transfer relationship is the driving transfer relationship between vehicles between domains determined based on the road network and the sample vehicle driving path.

22. The device according to claim 21, wherein The vehicle driving route recommendation request includes a driving starting point, a driving end point, and user characteristics of the vehicle user making the vehicle driving route recommendation request.

23. The device according to claim 21, wherein the route recommendation model is trained by a route recommendation model training device, wherein the route recommendation model training device comprises: An aggregation unit is used to aggregate named road sections with the same road name in the road network to obtain a named domain; Aggregating the unnamed road segments in the road network to obtain a virtual domain; a second determining unit, configured to determine a travel transfer relationship between vehicles between domains based on the road network and the obtained sample vehicle travel paths traveling on the road network, wherein the domains include the named domains and the virtual domains; A training unit is used to train a path recommendation model based on the driving transfer relationship and the sample vehicle driving path, wherein the path recommendation model is used to recommend a vehicle driving path.

24. The device according to claim 23, wherein The training unit comprises: A second generating subunit is configured to generate an undirected graph corresponding to the road network according to the driving transfer relationship; The training subunit is used to train the route recommendation model based on the undirected graph, the driving transfer relationship and the sample vehicle driving path.

25. The apparatus according to claim 24, further comprising: A second acquiring unit is configured to acquire the road name corresponding to each named road section from the road network; A construction unit, used to construct a mapping relationship between named road sections and road names; Furthermore, the training subunit is used to train the path recommendation model based on the undirected graph, the driving transfer relationship, the mapping relationship, and the sample vehicle driving path.

26. The device according to claim 25, wherein The training subunit comprises: a determination module, configured to determine a sequence of segment trajectories of a vehicle traveling on each segment in the road network based on the sample vehicle travel paths; a conversion module, configured to convert the road segment trajectory sequence into a domain trajectory sequence according to the undirected graph and the mapping relationship; A training module is used to train the path recommendation model according to the road segment trajectory sequence, the driving transfer relationship and the domain trajectory sequence.

27. The device according to claim 26, wherein The conversion module includes: a determination submodule, configured to determine, for each road segment in the road segment trajectory sequence, a domain corresponding to each road segment according to the mapping relationship; An acquisition submodule, configured to acquire, based on the undirected graph, connectivity relationships between domains corresponding to the determined road sections; A generation submodule is used to generate the domain trajectory sequence according to the connectivity relationship.

28. The device according to claim 26 or 27, wherein The road segment trajectory sequence has a trajectory starting point and a trajectory end point; the training module is used to train the path recommendation model according to the trajectory starting point, the trajectory end point, the driving transfer relationship and the domain trajectory sequence.

29. The apparatus according to claim 28, further comprising: a third acquiring unit, configured to acquire user characteristics of vehicle users corresponding to the road segment trajectory sequence; Furthermore, the training module is used to train the path recommendation model based on the user characteristics, the trajectory starting point, the trajectory end point, the driving transfer relationship, and the domain trajectory sequence.

30. The apparatus according to claim 29, wherein The training module includes: a splicing submodule, configured to perform feature splicing processing on the user feature, the trajectory starting point, the trajectory ending point, and the domain trajectory sequence to obtain a splicing feature; The training submodule is used to train the route recommendation model based on the driving transfer relationship and the splicing features.

31. The device according to any one of claims 23-27, 29-30, wherein The second determining unit includes: an extraction subunit, configured to extract connection relationships between domains in the road network; The second determining subunit is configured to determine the driving transfer relationship according to the connection relationship and the sample vehicle driving path.

32. The device according to any one of claims 23-27, 29-30, wherein There are multiple unnamed road sections; the aggregation unit includes: An acquisition subunit, configured to acquire an undirected graph corresponding to the road network; An exploration subunit, configured to use any unnamed road segment as an exploration starting point until a named road segment is explored from the undirected graph; The third determining subunit is configured to determine an area between any unnamed road segment and the explored named road segment as a virtual domain.

33. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 16.

34. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-16.

35. A computer program product comprising a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 16.

Images