Method and apparatus for path switching

Abstract

According to an embodiment of the present disclosure, a method, device, electronic device, computer storage medium and computer program product for path switching are provided. The method described herein includes: presenting a first path associated with a trip, the first path including a candidate path before the start of the trip or a current path during the trip; providing a risk reminder associated with the first path, the risk reminder is used to indicate a risky section in the first path, wherein the determination of the risky section is based at least on the road surface condition indicated by the road image; and providing a path switching entrance, the path switching entrance is used to trigger the provision of a second path and / or associate the second path with the trip, and the second path avoids the risky section. Based on the above method, the embodiment of the present disclosure can identify the risky section in the current route before the start of the trip or during the trip, so as to guide passengers to view it, and provide passengers with a quick route switching entrance to bypass the risky section.

Description

Technical Field

[0001] Example embodiments of the present disclosure generally relate to the field of computers, and more particularly, to methods, apparatuses, electronic devices, computer storage media, and computer program products for path switching. Background Art

[0002] At present, the information in the real world is updated quickly and has many dimensions. For transportation services, it is necessary to consider the special properties of the real environment when planning routes. For example, the actual passable width of the road, the muddiness of the road surface, and other factors that affect the traffic capacity.

[0003] In travel services (for example, online ride-hailing services), if drivers and passengers do not understand the overall traffic capacity of the travel route, it may result in the driver and passengers being unable to make timely decisions on whether to avoid risky sections of the road, thereby affecting the driver and passengers' travel experience. Summary of the invention

[0004] In a first aspect of the present disclosure, a method for route switching is provided. The method comprises: presenting a first route associated with a trip, the first route comprising a candidate route before the start of the trip or a current route during the trip; providing a risk reminder associated with the first route, the risk reminder being used to indicate a risky section in the first route, wherein the determination of the risky section is based at least on a road surface condition indicated by a road image; and providing a route switching entry, the route switching entry being used to trigger the provision of a second route and / or associate the second route with the trip, the second route avoiding the risky section.

[0005] In a second aspect of the present disclosure, a device for path switching is provided. The device includes: a path presentation module, configured to present a first path associated with a trip, the first path including a candidate path before the start of the trip or a current path during the trip; a risk provision module, configured to provide a risk reminder associated with the first path, the risk reminder is used to indicate a risky section in the first path, wherein the determination of the risky section is based at least on the road surface condition indicated by the road image; and an entry provision module, configured to provide a path switching entry, the path switching entry is used to trigger the provision of a second path and / or associate the second path with the trip, and the second path avoids the risky section.

[0006] In a third aspect of the present disclosure, an electronic device is provided, comprising: a memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to the first aspect of the present disclosure.

[0007] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which one or more computer instructions are stored, wherein the one or more computer instructions are executed by a processor to implement the method according to the first aspect of the present disclosure.

[0008] In a fifth aspect of the present disclosure, a computer program product is provided, comprising computer executable instructions, wherein the computer executable instructions implement the method according to the first aspect of the present disclosure when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0010] Figure 1 A schematic diagram showing an example environment in which embodiments according to the present disclosure may be implemented;

[0011] FIG. 2A to FIG. 2E A schematic diagram showing an example interface provided by a first terminal according to some embodiments of the present disclosure;

[0012] FIG. 3A to FIG. 3D A schematic diagram showing an example interface provided by a first terminal according to some other embodiments of the present disclosure;

[0013] Figure 4A and Figure 4B A schematic diagram showing an example interface provided by a second terminal according to some embodiments of the present disclosure;

[0014] Figure 5A and Figure 5B A schematic diagram showing a process of determining a risky road section according to some embodiments of the present disclosure is shown;

[0015] Figure 6 A schematic diagram showing an example process for path switching according to some embodiments of the present disclosure is shown;

[0016] Figure 7 A schematic structural block diagram of an apparatus for path switching according to some embodiments of the present disclosure is shown; and

[0017] Figure 8 A block diagram of an electronic device capable of implementing various embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0020] As discussed above, in travel scenarios such as online car-hailing, the route taken will greatly affect the quality of service provided. For example, some inappropriate routes may cause congestion and fail to reach the destination in time. Alternatively, some inappropriate routes may pass through some sections with poor road conditions, such as muddy sections, flooded sections, potholes, etc., which will greatly affect the passengers' riding experience. Therefore, a reasonable route is an important part of travel services.

[0021] In traditional travel services, the service recipients (e.g., passengers of online ride-hailing services) and service providers (e.g., passengers of online ride-hailing services) are usually unable to obtain risk alerts associated with recommended routes. This may result in drivers and passengers being unable to make timely decisions on whether to avoid risky sections, which in turn affects the driver and passenger travel experience.

[0022] A more serious situation is that due to the lack of information on risky sections, if the service provider enters a risky section during the travel service, it is difficult to guarantee the personal safety of the service recipient and the service provider, which may cause very serious consequences (for example, a traffic accident causing personal injury or vehicle damage).

[0023] In view of this, an embodiment of the present disclosure provides a solution for path switching. According to the solution, first, at a first terminal (for example, a mobile device of a passenger of an online car-hailing service), a candidate path before the start of the trip or a current path during the trip is presented as a first path and a risk reminder associated with the first path is provided. Such a risk reminder may indicate a risky section in the first path. The risky section may be determined based at least on the road surface condition indicated by the road image.

[0024] Furthermore, the first terminal provides a path switching entry for the provider of travel services (e.g., a passenger of a ride-hailing service). The path switching entry can trigger the provision of a second path that avoids the risk section, or the second path is associated with the current one.

[0025] Based on the above method, the embodiment of the present disclosure can support providing the acquirer of travel services with risk reminders related to the current trip through a terminal (e.g., a mobile device), so that the acquirer of travel services can make a decision on whether to avoid risky sections. In this way, the embodiment of the present disclosure can, on the one hand, enable the acquirer of travel services to decide on the desired driving route by themselves; on the other hand, it can also improve the safety of travel services.

[0026] Various example implementations of the scheme will be described in detail below with reference to the accompanying drawings.

[0027] Example Environment

[0028] See first Figure 1 , which schematically shows a diagram of an environment 100 in which an application according to an exemplary implementation of the present disclosure may be used.

[0029] Applications that enable users to call for vehicle services online are known and may be referred to as “travel service applications”, “ride-hailing software” or “ride-hailing applications”. Figure 1 As shown, the environment 100 includes a first terminal 110, which is configured to present Figure 1 The user interface 140 shown. In the user interface 140, the first terminal 110 can present a road risk reminder 130 related to a candidate route before the start of a trip or a current route during a trip to a user (e.g., a passenger). In some embodiments, the first terminal 110 can be, for example, an electronic device associated with a party acquiring travel services, examples of which can include but are not limited to: a smartphone, a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, a wearable device, etc.

[0030] As will be described in detail below, the first terminal 110 can, for example, inform the acquirer of the travel service of the information of the risky road section by presenting the risk reminder 130. In some embodiments, the first terminal 110 can generate the target route 150 by receiving the acquirer of the travel service's selection of path switching.

[0031] Alternatively, the server 120 may determine a risky section in a candidate path before the start of the trip or in a current path during the trip, and send the determined risky section to the first terminal 110 to trigger the first terminal 110 to provide a corresponding risk reminder.

[0032] The detailed process of route management for a trip is described in further detail below.

[0033] Sample route selection before your trip

[0034] In some embodiments, the embodiments of the present disclosure may, for example, support the acquirer of travel services to select the route of the trip before the trip begins. FIG. 2A to FIG. 2E , which shows schematic diagrams of example interfaces 200A to 200E provided by the first terminal 110 according to some embodiments of the present disclosure.

[0035] like Figure 2A As shown, the first terminal 110 can, for example, provide an interface 200A for the user through a travel application. The interface 200A can, for example, include input controls 205 and 210 for receiving the starting location (also called the departure point) and the end location (also called the destination) of the user's trip.

[0036] After receiving the starting point and the end point, the first terminal 110 may provide, for example, an interface 200B. The interface 200B is also referred to as a trip preview interface, and the user may select a desired route through the interface 200B.

[0037] In some embodiments, interface 200B may be presented before the travel service platform successfully assigns a service provider to the user. For example, the online car-hailing platform has not yet successfully assigned a driver to provide services to the user.

[0038] Alternatively, interface 200B may be provided after the travel service platform has successfully assigned a service provider to the user and before the service provider has started the trip. For example, the online car-hailing platform has assigned a driver to provide services to the user, but the driver is still on the way to pick up the user. Alternatively, the trip may also be a scheduled future trip that has not yet started.

[0039] like Figure 2B As shown, the interface 200B may present, for example, one or more recommended routes from the departure point 215 to the destination 220. Such a recommended route may be automatically generated by a travel platform, for example.

[0040] Additionally, the interface 200B may also include a route selection control 230, for example, for the user to select a corresponding recommended route. Figure 2B As shown, the route selection control 230 may include, for example, corresponding to three route types: "Current Route", "Shorter Time" and "Less Cost". After the user clicks the corresponding route selection control 230, the first terminal 110 may, for example, change the graphical representation of the route 235 corresponding to the route selection control 230, for example, the color, brightness, thickness, etc. of the route may be changed to indicate that it is selected by the user.

[0041] Interface 220B may provide risk reminders associated with candidate routes. The risk reminder may, for example, indicate a risky section in the candidate route. In some embodiments, a risk indicator corresponding to the risky section may be presented in interface 220B. For example, an indicator 250 indicating a construction section is presented on candidate route 240. An indicator 255 indicating a slippery road surface is presented on candidate route 235. In some other embodiments, the risky section may be presented in a preset style. For example, a section 260 indicating a rugged and potholed road surface is presented on candidate route 240. In addition, in some embodiments, interface 220B may also present an image corresponding to the risky section. Such an image may, for example, be acquired by a vehicle for providing travel services using an on-board image capture device.

[0042] In some embodiments, the first terminal 110 may, for example, present an interface element corresponding to the risk reminder on the interface 200B. The interface element may, for example, include a sign 250 indicating a construction section on the candidate route 240 described above, a sign 255 indicating a slippery road surface on the candidate route 235, and a road section 260 indicating a rough and pothole-prone road surface on the candidate route 240.

[0043] In some other embodiments, the first terminal 110 may provide the risk reminder to the acquirer of the travel service through a voice message. For example, the first terminal 110 may voice broadcast the information of the risky sections on the candidate route to the acquirer of the travel service (e.g., a passenger) during the presentation of the interface 200B. Alternatively or additionally, the first terminal 110 may trigger a call from the travel application to the acquirer of the travel service, thereby informing it of the information about the risky sections on the candidate route.

[0044] In some other embodiments, the first terminal 110 may also provide the risk reminder to the acquirer of the travel service via a text message. For example, the first terminal 110 may trigger the travel application to send an instant message to the acquirer of the travel service to inform it of information related to the risky sections on the candidate route. Figure 2B As shown, the first terminal 110 may also present a text message 265 about risk reminder through the interface 220B. The text message about risk reminder may be presented to the acquirer of the travel service through a pop-up window.

[0045] After presenting the risk reminder associated with the candidate route or during presenting the risk reminder associated with the candidate route, the first terminal 110 provides a path switching entry for the acquirer of the travel service. The path switching entry can, for example, trigger the provision of a candidate route that can avoid risky sections to the acquirer of the travel service.

[0046] like Figure 2CAs shown, in some embodiments, the first terminal 110 may present a control 270 for providing a path switching entrance to the acquirer of the travel service on the interface 200C. The control 270 may be provided through a pop-up window. For example, if the first terminal 110 determines that the risk level corresponding to the risk route is greater than a threshold, the path switching entrance is provided. It should be understood that any appropriate static threshold (e.g., a constant value) or dynamic threshold (e.g., a dynamic value that changes adaptively according to different scenarios) may be set according to actual needs as a basis for determining whether to provide a path switching entrance. In some other embodiments, the path switching entrance may also be a candidate route that does not have a risk section currently presented on the interface. For example, the acquirer of the travel service may select a route 245 that does not have a risk section (e.g., a risk route identifier is not presented) on the interface 200C.

[0047] In addition, if the first terminal 110 determines that the risk level corresponding to the risk route does not exceed the threshold, the first terminal 110 may present additional information about the risk section after receiving the selection for the risk reminder. Figure 2D As shown, if the first terminal 110 receives a request from the acquirer of the travel service to view the risk reminder in the candidate route (the sign 255 indicating the slippery road surface on the candidate route 235), for example, after detecting the click of the sign 255 by the provider of the travel service, the first terminal 110 can use the interface 200D to add information to view the interface 275, which can provide additional information of the risk section corresponding to the risk reminder and a control for switching to the path switching entrance of other alternative routes. In some embodiments, the additional information viewing interface 275 can be presented in the form of a pop-up window. In some other embodiments, the additional information viewing interface can also be an independent interface different from the interface 200D. For example, more detailed additional information of the risk section corresponding to the risk reminder can be presented on the independent interface, such as an image corresponding to the risk section, a driving risk level passing through the risk section, and an estimated time passing through the risk section.

[0048] If the first terminal 110 receives a selection of a route switch from the acquisition party of the travel service via the route switch portal, the first terminal 110 may, for example, change the graphical representation of the route corresponding to the selection of the route switch, for example, change the color, brightness, thickness, etc. of the route to indicate that it is selected by the user. Figure 2EAs shown, the switched route 245 corresponding to the selection of the route switch by the acquirer of the travel service is presented as a solid line in the interface 200E. If the first terminal 110 receives confirmation of the switched route by the acquirer of the travel service (for example, it is detected that the acquirer of the travel service inputs confirmation for the route 245 via the control 280), the first terminal 110 can determine that the upcoming trip will be carried out on the switched route 245.

[0049] Through the scheme described above, the party acquiring the travel service can select the route to be taken in the upcoming trip based on the risk reminders related to the candidate routes before the trip begins. In this way, on the one hand, it can meet the personalized needs of passengers for travel services and thus improve the customer experience, and on the other hand, it can significantly reduce the safety hazards caused by risky sections.

[0050] Example route switching during a trip

[0051] In some embodiments, the embodiments of the present disclosure may, for example, support the acquirer of travel services to change the route of the trip during the trip. FIG. 3A to FIG. 3C , which shows schematic diagrams of example interfaces 300A to 300E provided by the first terminal 110 according to some embodiments of the present disclosure.

[0052] like Figure 3A As shown, during the trip, the first terminal 110 may provide, for example, an interface 300A. The interface 300A is also called a trip progress interface, and the user may, for example, obtain trip information about the current travel service through the interface 300A, such as the route, the congested road sections on the route, the current vehicle position, the distance to the destination, the estimated arrival time, etc. As shown in the interface 300A, the vehicle is currently traveling on the route 320 and traveling to the location point 340.

[0053] The first terminal 110 can provide a risk reminder associated with the current driving route to the acquirer of the travel service through the interface 300A. The risk reminder can, for example, indicate a risky section in the current driving route. In some embodiments, a risk mark corresponding to the risky section can be presented in the interface 300A. For example, a mark 325 indicating a slippery road surface is presented on the route 320. In some other embodiments, the risky section can be presented in a preset style. In addition, in some embodiments, the first terminal 100 can also present an image corresponding to the risky section.

[0054] In some embodiments, the first terminal 110 may, for example, present an interface element corresponding to the risk reminder on the interface 300A. The interface element may, for example, include a sign 325 indicating that the road surface is slippery on the route 320 described above.

[0055] In some other embodiments, the first terminal 110 may provide the risk reminder to the acquirer of the travel service through a voice message. For example, the first terminal 110 may voice broadcast the information of the risk section on the current driving route to the acquirer of the travel service during the process of presenting the interface 300A. Alternatively or additionally, the first terminal 110 may trigger a call from the travel application to the acquirer of the travel service, thereby informing it of the information about the risk section on the candidate route.

[0056] In some other embodiments, the first terminal 110 may also provide the risk reminder to the acquirer of the travel service via a text message. For example, the first terminal 110 may trigger the travel application to send an instant message to the acquirer of the travel service to inform it of information related to the risky sections on the candidate route. Figure 3A As shown, the first terminal 110 may also present a text message 335 about risk reminder through the interface 300A. The text message about risk reminder may be presented to the acquirer of the travel service through a pop-up window.

[0057] like Figure 3B As shown, in some embodiments, the first terminal 110 may present a control 345 for providing a path switching entrance for the acquirer of the travel service on the interface 300B. The control 345 may be provided through a pop-up window. For example, if the first terminal 110 determines that the risk level corresponding to the risky route is greater than a threshold, the path switching entrance is provided. In some other embodiments, the path switching entrance may also be a candidate route that does not have a risky section currently presented on the interface. For example, the acquirer of the travel service may select a route 330 that does not have a risky section (e.g., a risky route identifier is not presented) on the interface 300B.

[0058] In addition, if the first terminal 110 determines that the risk level corresponding to the risk route does not exceed the threshold, the first terminal 110 may present additional information about the risk section after receiving the selection for the risk reminder. Figure 3CAs shown, if the first terminal 110 receives a request from the acquirer of the travel service to view the risk reminder in the candidate route (the sign 325 indicating the slippery road surface on the route 320), for example, after detecting the click of the provider of the travel service on the sign 355, the first terminal 110 can use the interface 300C to add information to view the interface 350, which can provide additional information of the risk section corresponding to the risk reminder and a control for switching to the path switching entrance of other alternative routes. In some embodiments, the additional information viewing interface 350 can be presented in the form of a pop-up window. In some other embodiments, the additional information viewing interface can also be an independent interface different from the interface 300C. For example, more detailed additional information of the risk section corresponding to the risk reminder can be presented on the independent interface, such as an image corresponding to the risk section, a driving risk level passing through the risk section, and an estimated time passing through the risk section.

[0059] If the first terminal 110 receives a selection of a route switch from the acquisition party of the travel service via the route switch portal, the first terminal 110 may, for example, change the graphical representation of the route corresponding to the selection of the route switch, for example, change the color, brightness, thickness, etc. of the route to indicate that it is selected by the user. Figure 3D As shown, the switched route 330 corresponding to the selection of the route switch by the acquirer of the travel service is presented in a solid line in the interface 300D. If the first terminal 110 receives confirmation of the switched route by the acquirer of the travel service (for example, it is detected that the acquirer of the travel service inputs confirmation of the route 330 via the control 360), the first terminal 110 can determine that the current travel service will be performed on the switched route 330.

[0060] Through the solution described above, the party who obtains the travel service can be informed in real time of the risky sections that will be passed in the current driving route of the vehicle in which the party is riding. This solution is particularly useful for temporarily increasing risky routes due to factors such as bad weather or sudden road accidents. In this way, the party who obtains the travel service can avoid risky sections in a timely manner, thereby significantly improving the safety factor of the travel service.

[0061] As described above, the travel service acquisition policy can select a route for the upcoming trip before the trip starts or switch the current driving route during the trip, for example, by inputting a confirmation for route 245 via control 280 in interface 200E or inputting a confirmation for route 330 via control 360 in interface 300D. After the first terminal 110 receives the above confirmation for the target route, the first terminal 110 can cause the target route and / or route information related to the target route to be sent (for example, via the server 120) to the second terminal 405 associated with the trip.

[0062] After receiving the target route, the second terminal 405 may, for example, present a user interface to provide information about the target route selected by the acquirer of the travel service. For example, the second terminal 405 may graphically present information about the target route selected by the acquirer of the travel service through a map application or a navigation application.

[0063] In some embodiments, such route information can be used to guide the provider of travel services to provide travel services accordingly. For example, the navigation application of the second terminal 405 can provide navigation prompts corresponding to the route information to guide the provider of travel services to provide travel services according to the route selected by the acquirer of travel services.

[0064] Figure 4A and 4B Schematic diagrams showing example interfaces 400A and 400B provided by a second terminal 405 according to some embodiments of the present disclosure. Figure 4A and 4B As shown, the second terminal 405 may present route information 410 about the target route 245 through the interface 400A, and may present route information 430 about the target route 330 through the interface 400B.

[0065] like Figure 4A As shown, the second terminal 405 may be, for example, a mobile device of a travel service provider (eg, a driver of an online car-hailing service), which may graphically represent route information 410 through a map component.

[0066] like Figure 4A As shown, interface 400A may also include prompt information 420 to remind the party acquiring the travel service (e.g., a passenger of an online car-hailing service) that the route of the trip has been selected.

[0067] In some embodiments, Figure 4A As shown, the travel service provider may still be on the way to pick up passengers. After completing the passenger pick-up and starting the trip, the second terminal 405 can use the navigation component to guide the trip based on the target route 245.

[0068] like Figure 4B As shown, the second terminal 405 may be, for example, a mobile device of a travel service provider (eg, a driver of an online car-hailing service), which may graphically represent route information 430 through a map component.

[0069] like Figure 4B As shown, interface 400B may also include prompt information 440 to remind the party acquiring the travel service (e.g., a passenger of an online car-hailing service) that the route of the trip has been changed.

[0070] It should be understood that although Figure 4A and 4B In the example of , the second terminal 405 is shown as a smart phone. In some scenarios, the second terminal 405 may also include other appropriate electronic devices, such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, a wearable device, a car host, etc.

[0071] For example, the second terminal 405 may be a vehicle computer of a vehicle driven by a driver of an online car-hailing service, which may provide route information about a target route by using an in-vehicle central control display system and / or an in-vehicle head-up display system.

[0072] In some embodiments, in addition to providing route information through a graphical interface, the second terminal 405 can also provide route information through text or voice. For example, the second terminal 405 can be a car computer, which can broadcast route information about the target route through voice to prompt the user that the route has been customized or guide the trip to travel according to the target route.

[0073] Alternatively or additionally, at the end of or after the trip of the current travel service, the first terminal 110 and / or the second terminal 405 may also provide a reminder that the completed trip avoided risky sections.

[0074] Based on this approach, the embodiments of the present disclosure can enable the acquirer of travel services (e.g., passengers of online ride-hailing services) to conveniently select or change routes and safely and effectively intervene in the travel route of the trip. Thus, the embodiments of the present disclosure can improve the convenience and friendliness of travel services.

[0075] Identification of risky sections

[0076] In some embodiments of the present disclosure, the risky road section can be determined based on the road condition of the road section. The road condition can indicate the road paving level of the road section. The road paving level can indicate, for example, whether the road section is muddy, pothole-ridden, a dirt road, or an asphalt paved road.

[0077] In addition, optionally or additionally, the risky road section can be determined based on other road section information. For example, the road width of the road section, the congestion level of the road section, the restricted traffic status of the road section and / or the road scene of the road section, etc. The road scene of the road section can indicate the environmental information around the road section. For example, the road scene of the road section can indicate whether there are obstacles in the road section (such as motor vehicles and non-motor vehicles parked along the road, vendors trading along the road, etc.).

[0078] In some embodiments, the road condition may be determined based on a road image of a road segment. For example, a vehicle for providing travel services may use an image acquisition device to acquire a road image of a road to determine the road segment information of the corresponding road segment.

[0079] In some embodiments, the determination of the risky road section may be implemented in the server 120. In some other embodiments, the determination of the risky road section may also be implemented at other devices associated with the travel service (eg, the first terminal 110).

[0080] Alternatively or additionally, risky road sections may be predicted by adopting a neural network-based model. Figure 5A An exemplary architecture of a neural network model predicting risky road sections is shown.

[0081] like Figure 5A As shown, an exemplary architecture for predicting risky sections may include a first model 510 and a second model 520. The first model 510 is used to determine the trajectory characteristics of the reference travel trajectory. For example, the first model 510 can calculate the characteristics of the trajectory under the risk scenario by extracting the trajectory expert representation and the trajectory depth representation. The second model 520 is used to determine the image classification features associated with the reference travel trajectory. For example, a video image is acquired through an acquisition device, and scene classification is performed through a deep model, and the classification confidence is used as a feature input trajectory representation to assist trajectory discrimination.

[0082] An exemplary architecture for predicting risky road sections may include multimodal model result fusion 530. The trajectory model determined by the first model 510 and the image model determined by the second model 520 are further fused at the multimodal model result fusion 530. The result of the first model 510 can be used to judge the quality of the road section condition associated with the reference travel trajectory (e.g., whether the road is easy to travel), and the result of the second model 520 can be used to understand the road section scene associated with the reference travel trajectory.

[0083] For example, Figure 5AAs shown, the first model 510 can perform trajectory acquisition 511. Trajectory acquisition may include trajectory acquisition 512 for the same origin and destination (OD) and all trajectory acquisition 513. The acquired trajectories for the same OD can characterize the differences in user route patterns. For example, 99% of users choose route A with the same OD, while only a very small number of users choose B, and most of the users who choose route B have abnormal behaviors such as low driving speed, abnormal IMU sensor parameters, and U-turns, which means that there are unreasonable sections on route B. The above-mentioned abnormal trajectory behaviors can be characterized by a variety of trajectory features (i.e., expert trajectory characterization 514), which may include, for example, user detour ratio, U-turn probability, low speed ratio, monthly traffic, and types of poi within 100 meters.

[0084] The above expert trajectory representation 514 is a feature set based on industry experience, but it still has some shortcomings such as incomplete dimensional feature design and low dimensionality, so a deep representation with stronger generalization and fitting capabilities is performed for all trajectories 513 (non-same OD) passing through a certain road. These trajectories can be input into the trajBert pre-training model 515 of the pre-training model Bert based on the application of natural language processing. The trajectory pre-training model based on Bert can use massive trajectory data for self-supervised learning, so as to make full use of massive historical trajectory data.

[0085] In some embodiments, the server 120 may map a set of trajectory points of the first trajectory to a set of spatial regions in a plurality of preset spatial regions. For example, in the trajBert pre-trained model 515, the trajectory data may be defined in the form of a specified point sequence, and then each trajectory point may be mapped to a preset spatial region (also referred to as a tile region) at a different level using a tile mapping mechanism, and then each trajectory point may be represented using tile labels at different levels of the tile. Figure 5B An exemplary schematic diagram of a tile mapping mechanism in the trajBert pre-training model 515 according to an embodiment of the present disclosure is shown.

[0086] like Figure 5B As shown, trajectory data 541 defined as a sequence of specified points is used as input to the trajBert pre-trained model 515. Each point in the trajectory data 541 can be mapped to multiple layers 542 0-N For example, N may be 20.

[0087] The entire tile-mapped spatial area is divided into four equal-sized sub-areas in the first layer 5420, namely, sub-areas 543, 544, 545, and 546, and then each sub-area is further divided into four equal-sized sub-areas in the second layer 5421, and so on. In this way, the entire longitude and latitude space of the earth can be divided into layers, so that any longitude and latitude point can correspond to the area of ​​each layer. This representation is also called tile representation. For example, the first point 547 of the trajectory in the figure is numbered 0 in the first layer area and 3 in the second layer area. The same is true for other layers, so the tile representation corresponding to this point is T0 = [0,3,…].

[0088] In addition, in order to accurately represent the specific position of each point in the area, alternatively or additionally, an offset rate can be used to represent the specific position of each point. For example, the specific position of the first point 547 of the trajectory in the figure in the sub-area 543 of the first layer 5420 is (0.5, 0.5), and its specific position in the sub-area 5421 of the second layer is (0.5, 0.1), and so on for other levels. Therefore, these offset tuples are uniformly represented as vectors, and the offset representation S0 = [0.5, 0.5, 0.5, 0.1, ...] can be obtained. Therefore, each longitude and latitude point can be regarded as consisting of a tile representation and an offset representation.

[0089] For tile representation, we first use one-hot encoding O j Indicates that the latitude and longitude point is at the jth layer, and then through the following formula: D j =W j ·O j Get the dense vector representation of the point at the jth layer, where W j The parameter matrix that needs to be learned for the jth layer is d×4, where d represents the dimension of the dense vector. Finally, the dense vectors of all layers are concatenated to obtain the dense vector D corresponding to the tile representation. T =concat(D1, D2, ...), the dimension of this vector is 20d.

[0090] Next, for the trajectory point, it not only contains a latitude and longitude point, but also contains speed and direction. In some embodiments, the server 120 can also determine the speed information and direction information associated with each trajectory point. Accordingly, in order to better characterize the trajectory point, the server can process the floating point values ​​corresponding to the speed and direction as follows: the speed is normalized by the maximum speed; the direction takes its trigonometric function value. Then, the processed speed and direction values ​​are spliced ​​into the offset representation, so that the dimension of the offset representation vector becomes: 2*20+1+2=43, and the changed vector is denoted as D s At this point, we can get the tile representation vector D of each trajectory point Tand the offset represents the vector D s , and then use them to concatenate to get a unified vector representation D ST .

[0091] The representation of the entire trajectory can be learned through the trajBert pre-trained model 415. Specifically, a number of points in the trajectory point vector sequence are randomly masked, and the sequence is concatenated with the position encoding vector sequence to obtain a new trajectory point encoding vector sequence, and then the trajectory point encoding vector is further encoded through the trajBert pre-trained model 415 to obtain a new vector representation, which is the implicit representation of each trajectory point.

[0092] The parameters of the relevant models can be learned based on two types of pre-training tasks. The first type is the consistency classification task. This task first constructs positive and negative samples by randomly intercepting and matching, then takes the latent representation sequence of the samples as input, and finally designs a classification model to determine whether the sequence comes from the same trajectory. For example, given two trajectories "ABCDEFG" and "HIJKLMN", the trajectory sequences after random interception and matching may be "ABCDLMN", "ABCDEF", etc., then "ABCDLMN" is a negative sample, and "ABCDEF" is a positive sample. Therefore, the effect of the pre-training model can be improved by this artificially constructed supervised learning task. The second type is the partial trajectory point feature prediction task. This task first randomly replaces the vector representation of some trajectory points with default vectors to cover up the information of these points, then restores the covered information through the prediction model, and finally learns the parameters of the model through the difference between the predicted information and the original information.

[0093] Here we introduce the multi-head attention mechanism. The multi-head attention mechanism can contain multiple attention modules, each of which contains three inputs: query input Q, key input K and value input V. For example, if the sequence is "ABCD", then Q in the first multi-head attention is "A", K and V are "ABCD", Q in the second multi-head attention is "B", K and V are still "ABCD", and so on. It can be seen that Q in the fourth multi-head attention is "D", and K and V are still "ABCD". Among them, the size of attention is the similarity between elements, which is calculated by cosine similarity. The output can be obtained by splicing the multi-head attention. Each output is regarded as a latent vector encoding representation, and then the parameters of the entire pre-trained model are learned through two self-supervised pre-training tasks: consistent classification task and partial trajectory point feature prediction.

[0094] The purpose of the consistent classification task is to determine whether the input sequence comes from the same trajectory. Therefore, the first choice is to construct positive and negative samples, where the positive sample is the sub-trajectory sequence data from the same trajectory, and the negative sample is the sub-trajectory sequence data intercepted from two different trajectories and then spliced ​​together, thereby generating supervised training samples, and finally using the cross entropy loss function to measure the classification quality.

[0095] The purpose of partial trajectory point feature prediction is to predict the relevant attributes of randomly masked points, including their position, speed, direction and other information. For example, given the input "ABCD", and some points will be randomly masked, such as masking point "B", the input becomes "A*CD" (where * is the default value), the task is to predict "B", and the learning method is to supervise the training of the model by calculating the difference between the predicted value and the true value. If the predicted attribute is a discrete value, it is measured by the classification loss function (cross entropy), and if it is a continuous value, it is measured by the mean square error (MSE, MeanSquare Error).

[0096] By combining the two pre-training tasks, the pre-training representation model can be learned simultaneously to obtain its parameter values, and finally a trained pre-trained TrajBert model is obtained.

[0097] Re-reference Figure 5A , the second model 520 can perform image acquisition 521, and extract multi-scene samples from the acquired images 522. By analyzing the acquired images and / or the extracted multi-scene samples, risky sections and non-risky sections can be distinguished. The recognition results obtained by the second model 520 will be integrated with the trajectory features to make a final judgment on the risky section.

[0098] The definition of risky road sections is not simply divided by the width of the road, but is judged from multiple aspects of the road, such as whether the road paving material is flat and bumpy, whether there is damage or cracks on the road, whether there is water or snow on the road, whether the road is occupied, whether there are many pedestrians or non-motorized vehicles on the road, etc. It should be understood that the width of the road itself is also an important aspect.

[0099] In some embodiments, the server 120 may process a set of images (e.g., sample images) associated with the first trajectory using the second model to determine road attributes associated with the first trajectory. In some embodiments, the road attributes indicate at least one of the following: road width, road pavement quality, and road scene.

[0100] In some embodiments, the width of a road can indicate the actual width of the road. Wider roads have better traffic capacity, while narrower roads have poor traffic capacity and are prone to becoming risky sections. The road paving quality indicates the pavement conditions. Asphalt roads are relatively flat and provide a good driving experience, while dirt roads, muddy roads, and damaged roads provide a poor driving experience and are prone to becoming risky sections. The road scene indicates the surrounding scenes of the road, such as expressways, rural roads, alleys, markets, etc. For some scenes, such as alleys and markets, driving is easily disturbed by surrounding pedestrians and non-motorized vehicles, making driving difficult and easily becoming risky sections.

[0101] After determining the definition of the dimensional standard, the model can be used to distinguish or detect various categories. In the selection of the model, a classification model can be used to distinguish several different categories. For example, the road width can be divided into three categories: "extremely narrow", "narrow", and "normal". Use the classification model to distinguish the three categories, and classify one or more of them as risky sections. At the same time, certain specific targets can also be detected by the detection model for final judgment. For example, in some embodiments, road damage and cracks can be detected to determine the quality of road paving. Alternatively or additionally, pedestrians and non-motor vehicles can also be detected to determine whether there is a lot of interference. In actual implementation, the above three dimensions can be classified using a classification model to determine risky sections. For example, the classification model can be implemented using SwinTransformer (Swin Transformer: Hierarchical Vision Transformer using ShiftedWindows).

[0102] For example, training data used by the model can be obtained through manual annotation. For example, a certain amount of data images are collected, and the images are manually classified into different categories according to the definition, or different detection targets are annotated in the images, and the annotated images are used for model training. When in use, the images are input into the trained model, the images are classified or detected, and the classification and detection results are reported to the model.

[0103] The result is output to the subsequent trajectory for fusion.

[0104] In some embodiments, the output results based on the SwinTransformer multi-scene model training prediction 524 may include image features extracted by the model. In some other embodiments, the output results based on the SwinTransformer multi-scene model training prediction 524 may also include the confidence of each category of the classification model, or include the number and confidence of a certain target of the detection model. In practical applications, the confidence of each category of the classification model can be used as the output based on the SwinTransformer multi-scene model training.

[0105] In this way, the trajectory features output by the first model 510 can be used to characterize the rationality of the trajectory. The image classification features output by the second model 520 can characterize the driving conditions of the road. Combining the two, the risky road sections in the trajectory can be comprehensively determined.

[0106] After obtaining the expert trajectory representation 514 and the TrajBert trajectory representation 515 in the first model 510, the output result of the SwinTransformer multi-scene model training prediction 524, i.e., the image scene confidence, can be integrated with the expert trajectory representation 514 and the TrajBert trajectory representation 515 to construct a tree model training prediction 516 (e.g., a gradient boosting decision tree) for model training and discrimination for judging the quality of road section conditions.

[0107] The output results of the tree model training prediction 516 and the SwinTransformer multi-scenario model training prediction 524 may be further fused at the multimodal model result fusion 530 to determine whether there is a risk in the road section associated with the reference travel trajectory.

[0108] According to the solution disclosed in the present invention, the above algorithm can realize the determination of risky sections, so that before or during the journey, the risky sections in the current route can be identified and the risk reminder about the risky sections can be provided to the acquirer of the travel service, so that the acquirer of the travel service can make decisions on the route of the travel service to avoid the risky sections. In this way, the customer experience of the travel service can be significantly improved, and the safety of the providers and acquirers of the travel service can be further guaranteed.

[0109] Example Process

[0110] Figure 6 FIG. 6 is a flow chart showing a process 600 for path switching according to various embodiments of the present disclosure. The process 600 may be performed as follows: Figure 1 The first terminal 110 is shown as an implementation.

[0111] like Figure 6 As shown, in block 610, the first terminal 110 presents a first route associated with a trip. The first route includes a candidate route before the start of the trip or a current route during the trip.

[0112] In block 620 , the first terminal 110 provides a risk reminder associated with the first route, wherein the risk reminder is used to indicate a risky section in the first route, wherein the risky section is determined based at least on a road surface condition indicated by the road image.

[0113] In block 630, the first terminal 110 provides a route switching entry. The route switching entry is used to trigger provision of a second route that avoids the risky route segment and / or associate the second route with the trip.

[0114] In some embodiments, providing a risk reminder associated with the first path includes: presenting a risky road section in a preset style; presenting a risk identifier corresponding to the risky road section; or presenting an image corresponding to the risky road section.

[0115] In some embodiments, the road condition indicates at least a road pavement level of the road segment.

[0116] In some embodiments, the determination of the risky road section is also based on: the road width of the road section; the congestion level of the road section; the restricted traffic status of the road section; or the road scene of the road section, which indicates the environmental information around the road section.

[0117] In some embodiments, providing the path switching entry includes: in response to a risk level corresponding to the risky road section being greater than a threshold, providing the path switching entry while providing a risk reminder associated with the first path.

[0118] In some embodiments, providing a path switching entry includes: receiving a selection for a risk reminder in response to a risk level corresponding to a risky road section being less than or equal to a threshold; and providing an additional information viewing interface, the information viewing interface presenting additional information and a path switching entry for describing the road passability.

[0119] In some embodiments, the method further includes: receiving a selection for a route switching entry at a first terminal, the first terminal being associated with a demander of travel services; and enabling a second terminal to provide a reminder about switching the trip to a route associated with a second route, the second terminal being associated with a provider of travel services.

[0120] In some embodiments, providing risk reminders includes: the first terminal providing risk reminders using voice messages and / or text messages; or the first terminal presenting a path viewing interface, the path viewing interface including interface elements corresponding to the risk reminders.

[0121] In some embodiments, the method further includes: in response to the end of the trip, providing a reminder that the ended trip avoided the risky road segment.

[0122] In some embodiments, the risky road section is determined based on the following process: using a first model to determine a trajectory feature of a first trajectory; using a second model to determine an image classification feature associated with the first trajectory; and based on the trajectory feature and the image classification feature, determining the risky road section in the first trajectory.

[0123] Example devices and equipment

[0124] The embodiments of the present disclosure also provide corresponding devices for implementing the above methods or processes. Figure 7 A schematic structural block diagram of an apparatus 700 for path switching according to some embodiments of the present disclosure is shown.

[0125] like Figure 7 As shown, the device 700 includes a route presentation module 710 configured to present a first route associated with a trip, where the first route includes a candidate route before the trip starts or a current route during the trip.

[0126] The apparatus 700 further includes a risk providing module 720 configured to provide a risk reminder associated with the first path, the risk reminder being used to indicate a risky section in the first path, wherein the determination of the risky section is based at least on a road surface condition indicated by the road image.

[0127] The device 700 also includes an entry providing module 730, which is configured to provide a path switching entry, the path switching entry is used to trigger the provision of a second path and / or associate the second path with the trip, and the second path avoids risky sections.

[0128] In some embodiments, the risk providing module 720 is further configured to present the risky road section in a preset style; present the risk identification corresponding to the risky road section; or present an image corresponding to the risky road section.

[0129] In some embodiments, the road condition indicates at least a road pavement level of the road segment.

[0130] In some embodiments, the determination of the risky road section is also based on: the road width of the road section; the congestion level of the road section; the restricted traffic status of the road section; or the road scene of the road section, which indicates the environmental information around the road section.

[0131] In some embodiments, the entry providing module 730 is further configured to provide a path switching entry while providing a risk reminder associated with the first path in response to a risk level corresponding to the risky road section being greater than a threshold.

[0132] In some embodiments, the entry providing module 730 is also configured to receive a selection for a risk reminder in response to a risk level corresponding to a risky road section being less than or equal to a threshold; and to provide an additional information viewing interface, which presents additional information and a path switching entry for describing the road passability.

[0133] In some embodiments, the device 700 is also configured to receive a selection for a path switching entry at a first terminal, the first terminal being associated with a demander of travel services; and to enable a second terminal to provide a reminder about switching the trip to an associated second path, the second terminal being associated with a provider of travel services.

[0134] In some embodiments, the risk provision module 720 is also configured to provide risk reminders by the first terminal using voice messages and / or text messages; or to present a path viewing interface by the first terminal, and the path viewing interface includes interface elements corresponding to the risk reminders.

[0135] In some embodiments, the apparatus 700 is further configured to provide a reminder that the completed trip avoided risky sections in response to the trip ending.

[0136] In some embodiments, the risky road section is determined based on the following process: using a first model to determine a trajectory feature of a first trajectory; using a second model to determine an image classification feature associated with the first trajectory; and based on the trajectory feature and the image classification feature, determining the risky road section in the first trajectory.

[0137] In some embodiments, using the first model to determine the trajectory characteristics of the first trajectory includes: mapping a set of trajectory points of the first trajectory to a set of spatial regions among a plurality of preset spatial regions; generating a vector representation for the first trajectory based on index information of the set of spatial regions and offset information of a set of trajectory points within the set of spatial regions; and processing the vector representation using the first model to determine the trajectory characteristics of the first trajectory.

[0138] In some embodiments, generating the vector representation for the first trajectory includes: determining speed information and direction information associated with each trajectory point; and generating the vector representation for the first trajectory based on the index information, the offset information, the speed information, and the direction information.

[0139] In some embodiments, determining image classification features associated with the first trajectory using a second model includes: processing a set of images related to the first trajectory using the second model to determine road attributes associated with the first trajectory, the road attributes indicating at least one of the following: road width, road pavement quality, and road scene; and determining image classification features associated with the first trajectory based on the road attributes.

[0140] In some embodiments, the trajectory features indicate the rationality of the trajectory, and the image classification features indicate the driving conditions of the road.

[0141] In some embodiments, determining the risky road section in the first trajectory based on the trajectory features and the image classification features includes: using a decision tree model to process the trajectory features and the corresponding image classification features to determine the risky road section in the first trajectory.

[0142] The units included in the device 700 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units can be implemented using software and / or firmware, such as machine executable instructions stored on a storage medium. In addition to or as an alternative to machine executable instructions, some or all of the units in the device 700 can be implemented at least in part by one or more hardware logic components. As an example and not limitation, exemplary types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0143] Figure 8 8 shows a block diagram of an electronic device / server 800 in which one or more embodiments of the present disclosure may be implemented. It should be understood that Figure 8 The electronic device / server 800 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein.

[0144] like Figure 8 As shown, the electronic device / server 800 is in the form of a general electronic device. The components of the electronic device / server 800 may include, but are not limited to, one or more processors or processing units 810, a memory 820, a storage device 830, one or more communication units 840, one or more input devices 850, and one or more output devices 860. The processing unit 810 may be an actual or virtual processor and is capable of performing various processes according to a program stored in the memory 820. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capabilities of the electronic device / server 800.

[0145] The electronic device / server 800 typically includes a plurality of computer storage media. Such media can be any available media accessible to the electronic device / server 800, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 820 can be a volatile memory (e.g., register, cache, random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory) or some combination thereof. The storage device 830 can be a removable or non-removable medium, and can include a machine-readable medium, such as a flash drive, a disk, or any other medium, which can be used to store information and / or data (e.g., training data for training) and can be accessed within the electronic device / server 800.

[0146] The electronic device / server 800 may further include additional removable / non-removable, volatile / non-volatile storage media. Figure 8 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. The memory 820 may include a computer program product 825 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.

[0147] The communication unit 840 implements communication with other electronic devices through a communication medium. Additionally, the functions of the components of the electronic device / server 800 can be implemented with a single computing cluster or multiple computing machines that can communicate through a communication connection. Therefore, the electronic device / server 800 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.

[0148] The input device 850 may be one or more input devices, such as a mouse, a keyboard, a tracking ball, etc. The output device 860 may be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device / server 800 may also communicate with one or more external devices (not shown) through the communication unit 840 as needed, such as a storage device, a display device, etc., communicate with one or more devices that allow a user to interact with the electronic device / server 800, or communicate with any device that allows the electronic device / server 800 to communicate with one or more other electronic devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).

[0149] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which one or more computer instructions are stored, wherein the one or more computer instructions are executed by a processor to implement the method described above.

[0150] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0151] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0152] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0153] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0154] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The selection of terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the marketplace, or to enable other persons of ordinary skill in the art to understand the implementations disclosed herein.

Claims

1. A method for path switching, comprising: presenting a first route associated with a trip, the first route comprising a candidate route before the trip begins or a current route during the trip; providing a risk reminder associated with the first path, the risk reminder being used to indicate a risky road section in the first path, wherein the determination of the risky road section is based at least on a road surface condition indicated by a road image; as well as Providing a path switching entry, the path switching entry is used to trigger the provision of a second path and / or associate the second path with the trip, the second path avoiding the risky section; The path switching entry provided includes: In response to a risk level corresponding to the risky road section being greater than a threshold, providing the path switching entry while providing the risk reminder associated with the first path; The risky sections are determined based on the following process: determining trajectory characteristics of the first trajectory using the first model; determining, using a second model, an image classification feature associated with the first trajectory; and The risky section in the first trajectory is determined based on the trajectory features and the image classification features.

2. The method of claim 1, wherein providing a risk alert associated with the first path comprises: Presenting the risky road section in a preset style; Presenting a risk indicator corresponding to the risky road section; or An image corresponding to the risky road section is presented. The method according to claim 1 , wherein the road surface condition at least indicates a road pavement level of the road segment.

4. The method according to claim 1, wherein the determination of the risky road section is further based on: Road width of the road section; The degree of congestion on the road section; the restricted access status of the road section; or A road scene of a road section, wherein the road scene indicates environmental information around the road section.

5. The method according to claim 1, wherein providing a path switching entry further comprises: In response to a risk level corresponding to the risky road section being less than or equal to a threshold, receiving a selection for the risk reminder; as well as An additional information viewing interface is provided, wherein the additional information for describing the passability of the road and the path switching entrance are presented.

6. The method according to claim 1, further comprising: Receiving a selection of the route switching entry at a first terminal, the first terminal being associated with a demander of a travel service; as well as The method enables a second terminal to provide a reminder about the trip switching to be associated with the second route, wherein the second terminal is associated with the provider of the travel service.

7. The method according to claim 6, wherein providing the risk reminder comprises: The first terminal provides the risk reminder by using a voice message and / or a text message; or The first terminal presents a path viewing interface, where the path viewing interface includes an interface element corresponding to the risk reminder.

8. The method according to claim 6, further comprising: In response to the trip ending, providing a reminder that the trip has ended and avoided the risky road section.

9. The method according to claim 1, wherein determining the trajectory characteristics of the first trajectory using the first model comprises: Mapping a set of trajectory points of the first trajectory to a set of spatial regions among a plurality of preset spatial regions; generating a vector representation for the first trajectory based on index information of the set of spatial regions and offset information of the set of trajectory points within the set of spatial regions; as well as The vector representation is processed using the first model to determine the trajectory characteristics of the first trajectory.

10. The method of claim 9, wherein generating a vector representation for the first trajectory comprises: determining speed information and direction information associated with each trajectory point; as well as The vector representation for the first trajectory is generated based on the index information, the offset information, the speed information, and the direction information.

11. The method of claim 1 , wherein determining, using a second model, an image classification feature associated with the first trajectory comprises: Processing a set of images associated with the first trajectory using the second model to determine a road attribute associated with the first trajectory, the road attribute indicating at least one of: road width, road pavement quality, and road scene; and Based on the road attributes, the image classification feature associated with the first trajectory is determined. 12 . The method according to claim 1 , wherein the trajectory feature indicates rationality of the trajectory, and the image classification feature indicates driving conditions of the road.

13. The method according to claim 1, wherein determining the risky road section in the first trajectory based on the trajectory feature and the image classification feature comprises: The trajectory features and the corresponding image classification features are processed using a decision tree model to determine the risky road section in the first trajectory.

14. A device for path switching, comprising: a route presentation module configured to present a first route associated with a trip, wherein the first route includes a candidate route before the trip starts or a current route during the trip; a risk providing module, configured to provide a risk reminder associated with the first path, the risk reminder being used to indicate a risky road section in the first path, wherein the determination of the risky road section is at least based on a road surface condition indicated by a road image; as well as an entry providing module, configured to provide a path switching entry, wherein the path switching entry is used to trigger the provision of a second path and / or associate the second path with the trip, wherein the second path avoids the risky section; The entry providing module is further configured to provide the path switching entry while providing the risk reminder associated with the first path in response to the risk level corresponding to the risk section being greater than a threshold; The risky sections are determined based on the following process: determining trajectory characteristics of the first trajectory using the first model; determining, using a second model, an image classification feature associated with the first track; as well as Based on the trajectory features and the image classification features, a risky road section in the first trajectory is determined.

15. An electronic device, comprising: Memory and processor; The memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to any one of claims 1 to 13.

16. A computer-readable storage medium having one or more computer instructions stored thereon, wherein the one or more computer instructions are executed by a processor to implement the method according to any one of claims 1 to 13.

17. A computer program product comprising computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 13.

Images