Speed prediction method, device and storage medium

Abstract

Embodiments of the present application provide a speed prediction method and device and a storage medium. After a road event occurs on an event link in a target road network, a detour path can be determined according to traveler information transparency and residual traffic capacity of the event link. The traveler information transparency is used to fully utilize the perception degree of travelers to the road event, which is conducive to more closely simulating the travel decision result of the travelers when encountering the road event. The residual traffic capacity of the event link is used as a prediction basis to improve the accuracy of the prediction result. The congestion spreading trend and the congestion dissipation trend of each link in the target road network are determined based on the relatively reliable prediction result of the traveler decision, and thus the influence of the road event on the target road network can be more truly reflected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traffic, in particular to a speed prediction method, device and storage medium. BACKGROUND

[0002] Urban traffic accidents have a significant impact on the short-term traffic capacity and traffic state of the road, and urban managers hope to quickly understand the impact of these traffic events on surrounding road segments and intersections, and to formulate appropriate management measures to eliminate congestion points based on this. In some traffic simulation methods, the prediction of congestion is usually based on the assumption that all people in the road network will make rational decisions and choose the shortest path under the influence of the event. However, in fact, in the case of a traffic accident in a short period of time, many people are not enough to grasp the traffic accident information and make new path planning and selection according to the current traffic state, so as to accurately predict the congestion situation after the road event and accurately predict the actual speed on the road after the event. Therefore, a new solution is needed. SUMMARY

[0003] The present application provides a speed prediction method, device and storage medium to predict the actual speed on the road after the occurrence of a road event.

[0004] The present application provides a speed prediction method, device and storage medium to predict the actual speed on the road after the occurrence of a road event.

[0005] Optionally, the step of determining the detour path according to the traveler information transparency and the residual traffic capacity of the event road segment comprises: determining an exponential distribution model corresponding to the detour position for bypassing the event road segment; the exponential distribution model is used to determine the detour starting point and the detour arrival point from the K-order neighborhood nodes of the event road segment; determining the value of the shape coefficient in the exponential distribution model according to the traveler information transparency and the residual traffic capacity, so as to determine the distribution characteristics of the detour position in the exponential distribution model according to the value of the shape coefficient; sampling the exponential distribution model to obtain at least one detour starting point and at least one detour arrival point; and determining the detour path according to the at least one detour starting point and the at least one detour arrival point.

[0006] Optionally, the shape coefficient is positively correlated with the traveler information transparency and negatively correlated with the residual traffic capacity.

[0007] Optionally, the determining the detour path according to the at least one detour starting point and the at least one detour ending point comprises: determining a plurality of candidate detour paths according to the at least one detour starting point and the at least one detour ending point; sorting the plurality of candidate detour paths according to a path quality index; the path quality index comprises a detour distance ratio and / or a path similarity; and selecting the detour path from the plurality of candidate detour paths according to a result of the sorting.

[0008] Optionally, the determining the traffic of each road segment in the target road network according to the detour path comprises: determining a traffic distribution ratio according to the residual traffic capacity of the event road segment; and distributing part of the traffic on the original path that needs to pass through the event road segment to the detour path according to the traffic distribution ratio, to update the traffic of each road segment in the target road network.

[0009] Optionally, the determining the congestion spread coefficient of each road segment in the target road network after the occurrence of the road event according to the traffic of each road segment in the target road network comprises: for a first road segment in the target road network, obtaining traffic entering the event road segment from the first road segment as first traffic, and obtaining traffic of the event road segment as second traffic; and determining a congestion spread coefficient on the first road segment that changes over time according to a ratio of the first traffic to the second traffic and a time-varying factor; the time-varying factor is determined according to the residual traffic capacity and the traveler information transparency that changes over time.

[0010] Optionally, the determining the congestion spread coefficient of each road segment in the target road network after the occurrence of the road event according to the traffic of each road segment in the target road network comprises: for a first road segment in the target road network, obtaining traffic entering the event road segment from the first road segment and flowing out of the event road segment as third traffic, and obtaining traffic of the event road segment as second traffic; and determining a congestion spread coefficient on the first road segment that changes over time according to a ratio of the third traffic to the second traffic and a time-varying factor; the time-varying factor is determined according to the residual traffic capacity and the traveler information transparency that changes over time.

[0011] Optionally, under the congestion spread coefficient and the congestion dissipation coefficient of each road segment in the target road network, the actual speed of each road segment in the target road network after the occurrence of the road event is predicted according to the historical speed of each road segment in the target road network, comprising: using a speed prediction model, predicting the event perception time length of each road segment in the target road network according to the congestion spread coefficient of each road segment in the target road network and the road congestion threshold value; and predicting the event processing time length of each road segment in the target road network according to the congestion dissipation coefficient of each road segment in the target road network and the road congestion threshold value; under a preset constraint condition, predicting the actual speed of each road segment in the target road network at different time after the occurrence of the road event according to the event perception time length of each road segment in the target road network, the event processing time length of each road segment in the target road network, the historical speed of the event road segment before the occurrence of the road event, the speed of the event road segment at different time after the occurrence of the road event, and the historical speed of each road segment in the target road network before the occurrence of the road event.

[0012] Optionally, the constraint condition comprises: the number of affected road segments, the number of congestion road segments and / or the number of recovered road segments at different time affected by the road event; wherein the number of affected road segments, the number of congestion road segments and / or the number of recovered road segments at any time is determined according to the congestion spread coefficient and the congestion dissipation coefficient of each road segment in the target road network at the time.

[0013] Optionally, before determining the flow of each road segment in the target road network according to the detour path, it further comprises: determining a first spatial range in a map according to the latitude and longitude coordinates of the road event and a specified position offset; determining a first sub-road network corresponding to the first spatial range in a road network; and determining a second sub-road network in which the neighbor nodes of the specified order of the road event are located; determining the union set of the first sub-road network and the second sub-road network to obtain a target road network, and determining the road segments affected by the road event in the target road network as each road segment in the target road network.

[0014] Optionally, determining each road segment affected by the road event in the target road network comprises: for any second road segment in the target road network, determining the flow correlation coefficient of the second road segment and the event road segment according to the flow of the event road segment and the flow of the event road segment flowing into the second road segment, and determining the second road segment as a road segment affected by the road event when the flow correlation coefficient is greater than a preset first threshold value; or determining the flow change proportion of the second road segment and the event road segment according to the flow change value of the second road segment after path detouring and the flow of the second road segment before path detouring, and determining the second road segment as a road segment affected by the road event when the flow change proportion is greater than a preset second threshold value.

[0015] An electronic device is also provided, including a memory and a processor; the memory is configured to store one or more computer instructions; the processor is configured to execute the one or more computer instructions to perform the steps in the method provided in the embodiments of the present application.

[0016] A computer readable storage medium storing a computer program is also provided, the computer program being executable by a processor to implement the steps in the method provided in the embodiments of the present application.

[0017] In the speed prediction method provided in the embodiments of the present application, after a road event occurs on an event link in a target road network, a detour path can be determined according to the traveler information transparency and the residual capacity of the event link. The traveler information transparency is used to fully utilize the perception degree of travelers on the road event, which is conducive to more realistically simulating the travel decision of the travelers when the road event occurs. The residual capacity of the event link is used to take the actual capacity of the event link as a prediction basis, thereby improving the accuracy of the prediction result. According to the detour path, the traffic of each link in the target road network can be determined, and according to the traffic of each link in the target road network, the congestion spreading trend and the congestion dissipation trend of each link in the target road network after the road event occurs can be determined. The congestion spreading trend and the congestion dissipation trend are determined based on the relatively reliable prediction result of the traveler decision, and thus can more truly reflect the influence of the road event on the target road network. Furthermore, under the congestion spreading trend and the congestion dissipation trend of each link in the target road network, the actual speed of each link in the target road network after the road event occurs, which is predicted according to the historical speed of each link in the target road network, can more accurately reflect the traffic state of the target road network under the influence of the road event. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which are included to provide a further understanding of the present application and are incorporated in and constitute a part of this application, illustrate embodiments of the present application and serve to explain the present application. In the drawings:

[0019] Figure 1 A flowchart of a speed prediction method provided for an exemplary embodiment of the present application is shown in FIG. 1;

[0020] Figure 2 A schematic diagram of a K-order neighborhood topological relationship of an event link provided for an exemplary embodiment of the present application is shown in FIG. 2;

[0021] Figure 3 A schematic diagram of a time-varying law of traveler information transparency provided for an exemplary embodiment of the present application is shown in FIG. 3;

[0022] Figure 4FIG. 1 is a schematic diagram of discrete sampling of a time range after a road event according to an example embodiment of the present application;

[0023] Figure 5 FIG. 4 is a schematic diagram of an electronic device according to an example embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described below in detail with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0025] The terms used in the embodiments of the present application are only for the purpose of describing the specific embodiments, and are not intended to limit the present application. The singular forms "a", "an" and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two, but does not exclude the case of including at least one.

[0026] It should be understood that the term "and / or" used herein is only to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " in this paper generally represents that the front and rear associated objects are a "or" relationship.

[0027] It should also be noted that the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, so that the product or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such product or system. Without more limitations, the element defined by the sentence "including a" does not exclude the presence of other identical elements in the product or system including the element.

[0028] In order to solve the technical problem that the existing scheme cannot accurately predict the congestion situation after the road event, in some embodiments of the present application, a solution is provided, and the technical solutions provided by the embodiments of the present application will be described in detail below with reference to the drawings.

[0029] Figure 1 FIG. 1 is a schematic diagram of discrete sampling of a time range after a road event according to an example embodiment of the present application; Figure 1

[0030] ​Step 101, determining a road event of an event road section in a target road network.

[0031] Step 102, determining a detour path according to traveler information transparency and residual traffic capacity of the event road section.

[0032] Step 103, determining traffic of each road section in the target road network according to the detour path.

[0033] Step 104, determining congestion spread coefficient and congestion dissipation coefficient of each road section in the target road network after the road event according to traffic of each road section in the target road network.

[0034] Step 105, predicting actual speed of each road section in the target road network after the road event according to historical speed of each road section in the target road network under congestion spread coefficient and congestion dissipation coefficient of each road section in the target road network.

[0035] The execution subject of the embodiment can be any electronic device, which can be an electronic device on the terminal side, such as a computer, a smart phone, etc., or an electronic device on the server side, such as a conventional server, a cloud server, a virtual machine, etc. The embodiment is not limited. The electronic device runs a prediction model, which can include one or more sub-models to realize congestion prediction after a road event.

[0036] The road event in the target road network can be an event that occurs in a certain road section in the target road network and has an impact on traffic, such as a traffic accident event, a natural disaster event (e.g., a landslide causing road section damage), a maintenance event (e.g., damaged road surface repair, road maintenance), etc.

[0037] After the road event occurs in the event road section, travelers on the event road section can switch paths to detour the event road section. As the event occurrence time elapses, the higher the traveler information transparency, the earlier the travelers make path switching decisions. The smaller the residual traffic capacity of the event road section, the greater the event impact, and the earlier the travelers make path switching decisions. Furthermore, the detour path at different time can be predicted based on traveler information transparency and residual traffic capacity of the event road section at different time.

[0038] The traveler information transparency is used to describe that the path decision rule of the traveler is affected by the traffic situation information currently mastered by the traveler. The greater the traveler information transparency is, the higher the overall perception degree of the traveler to the road event is, and the more the traveler can make or modify the travel decision in advance according to the road event. The traveler information transparency changes over time. Before a certain specified time after the event occurs, the traveler information transparency increases exponentially over time; after the specified time is reached, the traveler information transparency tends to be stable.

[0039] The residual passing capacity of the event road section is used to describe the capacity that can provide vehicle passing after the event road section is affected by the road event. The greater the influence of the road event is, the smaller the residual passing capacity of the event road section is; on the contrary, the greater the residual passing capacity of the event road section is. Based on the traveler information transparency and the residual passing capacity of the event road section at different times, the detour path is predicted, the perception degree of the traveler to the road event is fully considered, and the travel decision result of the traveler when the road event is encountered is simulated more close to the actual situation.

[0040] After the detour path of the traveler is predicted, the operation of the traveler detouring according to the detour path can be simulated. After the traveler detours, the traffic of each road section in the target road network will change. Therefore, the prediction model can determine the traffic of each road section in the target road network according to the detour path. The traffic of the road section refers to the number of vehicles passing through the road section per unit time.

[0041] After the traveler detours, the influence caused by the road event will present a point-to-surface propagation trend and can be slowly recovered to be unblocked under the management dredging operation. The prediction model can determine the congestion spreading trend and the congestion dissipation trend of each road section in the target road network after the road event occurs according to the traffic of each road section in the target road network. The congestion spreading trend is used to represent the probability that each road section in the event influence area is congested after the road event occurs. The congestion dissipation trend is used to represent the probability that each road section in the event influence area is recovered to be unblocked after the road event occurs. In this embodiment, the congestion spreading coefficient is used to represent the congestion spreading trend, and the congestion dissipation coefficient is used to represent the congestion dissipation trend. The optional calculation methods of the congestion spreading coefficient and the congestion dissipation coefficient will be introduced in subsequent embodiments, and details are not described herein.

[0042] After determining the congestion spreading trend and the congestion dissipation trend of each road segment, the prediction model can predict the actual speed of each road segment in the target road network after the road event according to the historical speed of each road segment in the target road network under the congestion spreading trend and the congestion dissipation trend. The historical speed of any road segment at any time can be the average speed of the road segment at multiple historical times, or the speed of the road segment at the same time on historical dates, which is not limited in the embodiment. In some embodiments, a model can be established according to the relationship between the actual speed of the road segment after being affected by the road event and the historical speed, the congestion spreading trend and the congestion dissipation trend. The model can be a function model based on a quadratic curve or a function model based on a neural network, which is not limited in the embodiment.

[0043] In the embodiment, after the event road segment in the target road network occurs a road event, a detour path can be determined according to the traveler information transparency and the residual capacity of the event road segment. The traveler information transparency is fully utilized based on the perception degree of the traveler to the road event, which is conducive to more realistically simulating the travel decision result of the traveler when encountering the road event. The actual capacity of the event road segment is taken as a prediction basis based on the residual capacity of the event road segment, which improves the accuracy of the prediction result. According to the detour path, the traffic of each road segment in the target road network can be determined, and the congestion spreading trend and the congestion dissipation trend of each road segment in the target road network after the road event can be determined according to the traffic of each road segment in the target road network. The congestion spreading trend and the congestion dissipation trend are determined based on the more reliable prediction result of the traveler decision, and thus can more truly reflect the influence of the target road network by the road event. Further, the actual speed of each road segment in the target road network after the road event predicted according to the historical speed of each road segment in the target road network under the congestion spreading trend and the congestion dissipation trend of each road segment in the target road network can more accurately reflect the traffic state of the target road network under the influence of the road event.

[0044] It should be noted that, in the above and the following embodiments of the present application, in order to reduce the calculation complexity of the model, the target road network refers to a sub-road network in the road network. The sub-road network can be determined according to the location of the road event. The following will be exemplarily described.

[0045] Optionally, the first spatial range can be determined in the map according to the latitude and longitude coordinates of the road event and a specified position offset, and a first sub-road network corresponding to the first spatial range in the road network can be determined. A second sub-road network in which the neighbor nodes of the specified order of the road event are located can be determined, and a union set of the first sub-road network and the second sub-road network can be determined to obtain the target road network. After obtaining the target road network, each road segment in the target road network affected by the road event can be determined.

[0046] Specifically, a spatial range M1 can be defined by offsetting the longitude and latitude coordinates of the event occurrence location by a specified number of degrees, and a sub-road network within the spatial range M1 can be extracted from the road network. In some embodiments, to reduce the risk that the defined spatial range M1 corresponds to a road network that is too sparse, a K-order neighbor node can be expanded from the event occurrence location to the road network to obtain a topological range M2, where K is a positive integer. The union set U(A, B) of the sub-road network within the spatial range M1 and the topological range M2 can be taken as the target road network M.

[0047] Optionally, after the target road network is determined, the affected road segments and the affected area of the event can be determined from the target road network. The affected road segments refer to road segments in the target road network that have a strong flow dependence relationship with the event road segment, and the area formed by the plurality of affected road segments is the affected area.

[0048] In some optional embodiments, the affected road segments can be filtered from the target road network by a flow correlation coefficient. Taking any road segment in the target road network as an example, the flow correlation coefficient of the road segment and the event road segment can be determined according to the flow of the event road segment and the flow of the event road segment flowing into the road segment. Optionally, the flow correlation coefficient of any road segment in the target road network and the event road segment can be determined according to the ratio of the traffic volume flowing from the event road segment to the road segment to the traffic volume of the event road segment. If the flow correlation coefficient is greater than a first threshold value, the road segment can be determined to be an affected road segment. For example, the flow correlation coefficient a1 of the road segment i and the event road segment (road segment j) in the target road network can be expressed as:

[0049]

[0050] where Volume(j→i) is the traffic volume flowing from the event road segment (road segment i) to the road segment i, and Volume(j) is the traffic volume of the event road segment (road segment j). When a1≥a0, the road segment i can be determined to be an affected road segment of the event, where a0 is a set threshold value that can be set according to experience.

[0051] In some optional embodiments, the calculation of the detour path and the path optimization operation based on the detour path can be performed according to a set period. The affected road segments can be filtered from the target road network according to the flow change ratio of each road segment before and after path optimization. Taking any road segment in the target road network as an example, the flow change ratio of the road segment and the event road segment can be determined according to the flow change value of the road segment after path detouring and the flow of the road segment before path detouring. Optionally, the flow change ratio of any road segment can be determined according to the ratio of the flow difference of the road segment before and after path optimization to the flow of the road segment before path optimization. If the flow change ratio is greater than a second threshold value, the road segment can be determined to be an affected road segment. For example, the flow change ratio a2 of the road segment i in the target road network can be expressed as:

[0052]

[0053] wherein, Volume pre (i) is the traffic volume of link i before route optimization; Volume post (i) is the traffic volume of link i after route optimization; ABS() represents calculating absolute value. When a2≥a0, it can be determined that link i is the influence link of the event. Based on this implementation, the risk of the calculated influence link being discontinuous in space can be reduced, and the risk of the link far away from the event link being determined in the influence area can be reduced.

[0054] In this embodiment, the prediction model can include a route optimization model, which is used to predict a target route (i.e., a detour route) that needs to be detoured due to the influence of the event link. The following will illustrate the optional ways of route optimization in combination with the route optimization model.

[0055] After determining the target road network, the influence link, and the influence area, the electronic device can predict a detour route in the target road network based on a route optimization model (M1). The prediction of the detour route is mainly used to predict the degree of perception of road events by travelers and simulate the travel choices made by travelers after perceiving the road events. The following will be illustrated exemplarily.

[0056] In the target road network, the relationship between the elements of the target road network can be represented by a topological graph, the nodes in the topological graph represent the endpoints of the links, and the lines in the topological graph represent the links. In the topological graph, the nodes that can be reached by passing through at least K edges from the endpoints of the event link are called K-order neighborhood nodes of the event link, which are simply referred to as k-order neighborhood nodes, as shown in Figure 2 .

[0057] In some exemplary embodiments, when determining the detour route according to the information transparency of the traveler and the residual traffic capacity of the event link, an exponential distribution model corresponding to the detour position can be determined, and the detour route can be determined according to the exponential distribution model. The exponential distribution model is used to determine the detour starting point and the detour arrival point from the K-order neighborhood nodes of the event link. The exponential distribution can be any one of a Gamma Distribution, a Beta Distribution, and a Dirichlet Distribution. According to the information transparency of the traveler and the residual traffic capacity, the value of the shape coefficient in the exponential distribution model can be determined, and the shape coefficient is used to represent the distribution characteristics of the detour position. Sampling the exponential distribution model can obtain at least one detour starting point and at least one detour arrival point.

[0058] In the path optimization model, an exponential distribution model can be used to predict the bypass departure point and the bypass arrival point and the adjacent order number where the bypass departure point and the bypass arrival point are located according to probability. Taking a Gamma distribution as an example, the probability density function p(x) of the Gamma distribution is as follows:

[0059]

[0060] where k is a shape coefficient, θ is a scale coefficient, x is a random variable, e is a natural constant, and Γ(k) is a Gamma function; the Gamma function Γ(k) can be expressed by the following formula:

[0061]

[0062] Based on the integral of the above probability density function, the Gamma distribution can be determined. By sampling the Gamma distribution, the bypass position can be obtained. Taking the bypass departure point in the bypass position as an example, when predicting the bypass departure point, it is assumed that the bypass departure point is located in the x-order adjacent area upstream of the event link, and x follows the Gamma distribution x ~ Gamma(k, θ) with parameters k and θ. Based on the sampler sampling the Gamma distribution, the adjacent order number where the bypass departure point is located can be obtained. According to the adjacent order number where the bypass departure point is located and the topology of the K-order adjacent area, the bypass departure point can be determined. The sampler can be a sampler based on the Monte Carlo sampling method.

[0063] In the Gamma distribution, the shape parameter k can be determined according to the law of the bypass decision made by the traveler. Generally, as the event occurrence time elapses, the higher the information transparency of the traveler, the earlier the traveler makes the path switching decision. The smaller the residual capacity of the event link, the greater the impact of the event, and the earlier the traveler makes the path switching decision. Further, the shape coefficient k in the Gamma distribution can be determined based on the information transparency of the traveler and the residual capacity of the event link, so that the Gamma distribution can express the distribution characteristics of the bypass position. The determination methods of the information transparency of the traveler and the residual capacity of the event link will be described below.

[0064] Alternatively, the information transparency τ of the traveler is used to describe the influence of the traffic situation information currently mastered by the traveler on the path selection behavior of the traveler. The information transparency of the traveler changes over time, and before a certain specific time after the event occurs, the information transparency of the traveler shows an exponential rising trend as time elapses; after reaching the specified time, the information transparency of the traveler tends to be stable. In some optional embodiments, the time-varying law corresponding to the information transparency of the traveler can be described by using the following time-varying model:

[0065]

[0066] wherein t is the time elapsed since the event, T is the total time of the event; m is an experience value, which can be taken as m = 2 generally; t max is the maximum influence time of the road event, which is different for different event types, and generally t max = T + 0.5; τ max is the maximum value of the traveler information transparency under the influence of the road event, which is generally taken as 1. The time-varying model corresponding to the traveler information transparency can present a form as shown in Figure 3 : before the t max moment after the event occurs, the traveler information transparency presents an exponential upward trend with time; after the t max moment, the traveler information transparency no longer changes.

[0067] Figure 3 The time-varying law of the traveler information transparency is described by a continuous function, in order to facilitate calculation, a plurality of anchor timestamps can be selected within the time range of the accident occurrence, and the traveler information transparency at a plurality of discrete time points is used to describe the change of the entire event occurrence period. As shown in Figure 4 , a plurality of discrete time points, such as t start , t mid and t end , can be selected within the time range of the accident occurrence, and the traveler information transparency at the discrete time points is used to participate in subsequent calculation.

[0068] Optionally, the residual traffic capacity r c of the event road segment is used to describe the residual capacity of the event road segment that can provide vehicle traffic after the event occurs. In some embodiments, the residual traffic capacity r c of the event road segment can be determined by the ratio of the residual available lane number of the event road segment after the event occurs to the total lane number of the event road segment. For example, the residual traffic capacity r c of the event road segment can be expressed by the following formula:

[0069]

[0070] wherein N La is the residual available lane number of the event road segment after the event occurs, and N Lt is the total lane number of the event road segment.

[0071] After the traveler information transparency and the residual traffic capacity of the event road segment are determined based on the above embodiments, the shape coefficient k in the Gamma distribution can be determined based on the traveler information transparency and the residual traffic capacity of the event road segment. In some optional embodiments, the shape coefficient k can be calculated in the following manner:

[0072]

[0073] wherein k0, τ0are hyperparameters, in some cases, k0= 2, τ0= 0.5 can be taken as empirical values. Wherein τ is the information transparency of the traveler, r c is the residual capacity of the event link.

[0074] As shown in formula 1-3, the shape coefficient k in the Gamma distribution is positively correlated with the information transparency τ of the traveler, that is, the higher the information transparency of the traveler, the larger the shape coefficient, and the peak value in the Gamma distribution curve is shifted to the right as a whole, and the distance from the detour starting point to the event link obtained by sampling the Gamma distribution is farther. As shown in formula 1-3, the shape coefficient k is negatively correlated with the residual capacity r c of the event link, that is, the greater the residual capacity of the first road after the event (the smaller the event impact), the smaller the shape coefficient k, and the peak value in the Gamma distribution curve is shifted to the left as a whole, and the distance from the detour position to the event link obtained by sampling the Gamma distribution is closer. Based on this implementation, by determining the shape parameter in the Gamma distribution based on the information transparency of the traveler and the residual capacity of the event link, the distribution characteristics of the detour position can be more accurately expressed by the Gamma distribution.

[0075] Similarly, when predicting the detour arrival point, it can be assumed that the detour arrival point is located in the x-th neighborhood downstream of the event link, and x is subject to the Gamma distribution x ~ Gamma(k, θ) with parameters k and θ. By sampling the Gamma distribution, the neighborhood order of the detour arrival point can be obtained. According to the neighborhood order of the detour arrival point and the topology of the K-th neighborhood, the detour arrival point can be determined.

[0076] By sampling the Gamma distribution corresponding to the detour starting point and the detour arrival point respectively, the neighborhood order of the detour starting point upstream and the neighborhood order of the detour arrival point downstream can be obtained. Wherein, the same order of the detour starting point or the detour arrival point obtained by sampling can be one or more, which is not limited in this embodiment. If multiple detour starting points are obtained by sampling, the neighborhood orders of the multiple detour starting points can be the same or different; if multiple detour arrival points are obtained by sampling, the neighborhood orders of the multiple detour arrival points can be the same or different.

[0077] After determining at least one detour starting point and at least one detour arrival point, the path optimization model can calculate the detour path according to the at least one detour starting point and the at least one detour arrival point.

[0078] In some optional Example AIn some embodiments, the path optimization model can sample the exponential distribution model (i.e., Gamma distribution) twice by the sampler to obtain the detour starting point n s and the detour ending point n t , respectively. The path optimization model can determine the detour path between the detour starting point n s and the detour ending point n t based on the path length and the topology of the target road network. For example, the path optimization model can calculate the shorter path between the detour starting point n s and the detour ending point n t based on the Dijkstra algorithm, the Floyd algorithm (also known as the insert point method), or the like, and select the shorter path as the detour path.

[0079] Optionally, the path optimization model can set a weight coefficient for each road segment in the target road network during the calculation of the detour path. For example, the path optimization model can set the weight coefficient for different road segments according to the road grade or the upper and lower limits of the high speed limit. Thus, the path optimization model can calculate the length of at least one path between the detour starting point n s and the detour ending point n t based on the weight coefficient and the length of each road segment, and select the shorter path as the detour path.

[0080] In some optional Example B embodiments, the path optimization model can predict multiple detour starting points (n s1 , n s2 , n s3 ,...) and multiple detour ending points (n t1 , n t2 , n t3 ,...) of the same order when performing the detour location prediction. Then, the path optimization model can determine multiple candidate detour paths based on the multiple detour starting points and the multiple detour ending points, and determine the optimal detour path between the multiple detour starting points and the multiple detour ending points by path screening.

[0081] Optionally, the path optimization model can sort the multiple candidate detour paths based on at least one path quality indicator, and select the detour path from the multiple candidate detour paths based on the sorting result. For example, the path optimization model can select the candidate detour path with a higher ranking or the candidate detour path with the first ranking.

[0082] Optionally, the indicator for measuring the optimal detour path can include at least the detour distance ratio and / or the path similarity. The following will be described in detail.

[0083] In an optionalExample B1 In some embodiments, the plurality of detour paths can be ranked according to the detour distance ratio, and the optimal detour path can be selected according to the ranking result. The detour distance ratio refers to the ratio of the length of the detour path to the length of the original path, which can be expressed by the following formula:

[0084]

[0085] where i represents the sequence number of the road segment included in the detoured path, and j represents the sequence number of the road segment included in the detoured path. i is the road segment i included in the detoured path, j is the road segment j included in the original path, and L represents the length. d is the detour distance ratio.

[0086] After calculating the detour distance ratio of the plurality of detour paths and the original path based on the above method, the plurality of detour paths can be ranked, and the optimal detour path can be selected according to the ranking result.

[0087] In an optional Example B2 embodiment, the plurality of detour paths can be ranked according to the path similarity, and the optimal detour path can be selected according to the ranking result. The path similarity refers to the similarity between the detour path and the original path.

[0088] For example, the path optimization model can use the cumulative distance in the dynamic time warping algorithm (DTW) as a measure of path similarity. The dynamic time warping algorithm assumes that the detour path and the original path each contain a certain number of key points, and calculates the cumulative distance of the time-warped key points. For details, refer to the following formula:

[0089] s s = γ(i, j) = d(q i , c j ) + min{γ(i-1, j-1), γ(i-1, j), γ(i, j-1)} Formula 2-7

[0090] where s s is the cumulative distance; γ() is the calculation function of the path similarity, d() is the distance calculation function; q i is the key point on road segment i, and c j is the key point on road segment j. The above formula 1-5 calculates the sum of the Euclidean distance between the key point q i and the key point c j and the cumulative distance of the smallest adjacent element that can reach the point q i in a recursive manner.

[0091] After the path similarity of the multiple detour paths and the original path is calculated based on the above manner, the multiple detour paths can be sorted, and a better detour path can be selected according to the sorting result.

[0092] In the above embodiment, the screened better path can be one or more, and the present embodiment is not limited.

[0093] After the detour path is predicted, it can be assumed that the travelers on the original path (the original path without detouring through the event section) switch paths through the detour path. A part of the traffic on the original path needs to be allocated to the detour path. The path optimization model can further calculate the traffic allocation after the path switching.

[0094] Optionally, the traffic retention ratio of the original path can be any specified target ratio, such as 50%, 60%, or other ratios. Optionally, considering the traffic conditions of the original path, the traffic retention ratio of the original path can be determined according to the remaining capacity r of the event section c determination. The path optimization model can determine the traffic allocation ratio according to the remaining capacity of the event section, and allocate part of the traffic on the original path that needs to pass through the event section to the detour path according to the traffic allocation ratio, to update the traffic of each section in the target road network. Assuming that there are n detour paths, the proportion of the traffic allocated from the original path to each detour path can be P:

[0095]

[0096] Based on the above embodiment, the traffic on each path after path optimization can be predicted, thereby facilitating the calculation of the congestion situation of the target road network after the event occurs.

[0097] After the traffic is redistributed, the road congestion coefficient model (M2) can be used to determine the congestion spread coefficient and the congestion dissipation coefficient of the road section level after the road event occurs, based on the redistributed traffic. The congestion spread coefficient is used to describe the congestion spread trend, and the congestion dissipation coefficient is used to describe the congestion dissipation trend.

[0098] In the present embodiment, the road congestion coefficient model can obtain the congestion spread coefficient and the congestion dissipation coefficient of the road section level at different time slices (t time). The input of the road congestion coefficient model can include: the event section (i.e. the event section), the historical path in the influence area of the event, the traffic of the influence section of the event, and the topology structure of the target road network; the output of the road congestion coefficient model can include: the congestion spread coefficient and the congestion dissipation coefficient.

[0099] The determination methods of the congestion spread coefficient and the congestion dissipation coefficient will be described in the following examples.

[0100] Taking the first road segment (i.e., any road segment) in the target road network as an example, the road congestion coefficient model can obtain the traffic flow from the first road segment to the event road segment as the first traffic flow, and obtain the traffic flow of the event road segment as the second traffic flow. The road congestion coefficient model can determine the congestion spread trend on the first road segment over time based on the ratio of the first traffic flow to the second traffic flow and a time-varying factor. The time-varying factor is determined based on the remaining capacity and the transparency of traveler information over time. For ease of description, t... n At any given time, the probability of congestion on road segment i due to road events on segment j can be expressed using the congestion spread coefficient. To express.

[0101] Optionally, changes in the traffic capacity of the affected road segment cause congestion upstream of the original route that needed to pass through the affected road segment, leading to congestion on the corresponding upstream road segment at different times. Therefore, the congestion spread coefficient can be determined based on the traffic flow on the original route that needed to pass through the affected road segment. As shown in the following formula:

[0102]

[0103] Where i represents the affected road segment, and j represents the event road segment. It refers to t n The remaining capacity on road segment j at time t0 is the initial time of the event, t n P represents the nth moment after the event occurs. in,i→j,t P represents the cumulative traffic flow from segment i to segment j up to time t; out,i→j,t R represents the cumulative outflow of traffic from segment i to segment j up to time t; i→j,t Let P represent the topological structure of the path from road segment i to road segment j at time t. in,i→j,t and P out,i→j,t It can be determined based on historical routes within the affected area and traffic flow on the affected road segments.

[0104] Formula 2-1 above can be simplified to the following formula:

[0105]

[0106] Where Volume(i→j) represents the traffic flow from road segment i to event road segment j (i.e., the first flow), and Volume(j) is the traffic flow of road segment j (i.e., the second flow). Their ratio represents the traffic flow correlation between the current road segment i and the event road segment j. ξ t ξ is a time-varying factor. tis negatively correlated with the information transparency of the traveler at time t. The greater the information transparency of the traveler at time t, the smaller the time-varying factor ξ t In some embodiments, the time-varying factor ξ t may be expressed by the following formula:

[0107]

[0108] wherein r c is the residual traffic capacity, τ t is the information transparency of the traveler at time t.

[0109] Based on this implementation, the information transparency of the traveler and the traffic volume of the event road section and the affected road section are comprehensively considered to determine the congestion spread coefficient after the event, and the decision information of the traveler is fully considered, so that the calculated congestion spread coefficient is more in line with the actual scene.

[0110] Continuing to take the first road section as an example, the road congestion coefficient model can obtain the traffic volume flowing into the event road section from the first road section and flowing out of the event road section as the third traffic volume, and obtain the traffic volume of the event road section as the second traffic volume; the road congestion coefficient model can determine the congestion dissipation trend of the first road section over time according to the ratio of the third traffic volume to the second traffic volume and the time-varying factor; wherein the time-varying factor is determined according to the residual traffic capacity and the information transparency of the traveler over time.

[0111] For ease of description, at time t n , the probability that the road section i recovers to be unblocked under the influence of the road event on the road section j can be expressed by the congestion dissipation coefficient In some embodiments, the congestion dissipation coefficient may be set as a fixed value according to experience. In other embodiments, the congestion dissipation coefficient may be determined based on the traffic volume on the path that originally needs to pass through the event road section, as shown in the following formula:

[0112]

[0113] wherein, is the residual traffic capacity of the road section j at time t n ; t0 is the initial time of the event, and t n is the n th time after the event. P in,i→j,t represents the traffic volume passing through the road section i to the road section j by time t; P out,i→j,t represents the cumulative traffic volume of the road section i flowing out of the road section j by time t. R i→j,t represents the topological structure of the path from the road section i to the road section j at time t.

[0114] wherein, if the road section i is affected by multiple event road sections, the congestion spread coefficient β of the road section i i,t The congestion spread coefficient of the road section i under the influence of multiple event road sections j can be an average value, i.e., Correspondingly, the congestion dissipation coefficient β of the road section i i,t The congestion dissipation coefficient of the road section i under the influence of multiple event road sections j can be an average value, i.e.,

[0115] Based on the foregoing embodiments, the congestion spread coefficient β of different road sections at different time is determined i,t and the congestion dissipation coefficient μ i,t Then, the event influence data at the regional level can be calculated. In this embodiment, the road network congestion constraint model (M3) can be used to calculate at least one of the number of affected road sections S(t), the number of congested road sections I(t) and the number of recovered road sections R(t) in the influence area at different times. The number of affected road sections, the number of congested road sections and / or the number of recovered road sections are used as constraint conditions for predicting the speed of each road section under the influence of the event, i.e., the speed of each road section predicted under the congestion spread trend and the congestion dissipation trend at different times needs to meet the number requirements of different types of road sections of the target road network at different times. The number of affected road sections, the number of congested road sections and the number of recovered road sections at any time are determined according to the congestion spread coefficient and the congestion dissipation coefficient of each road section in the target road network at that time. The calculation process of the above constraint conditions will be described exemplarily below.

[0116] In this embodiment, for the convenience of calculation, the continuous time after the event occurs is sampled to obtain multiple discrete times. Considering the discontinuity of the time granularity, the road network congestion constraint model can be described as follows:

[0117]

[0118] wherein, S(t) represents the number of affected road sections at time t, ΔS(t+1) represents the increment of the number of affected road sections at time t+1, S(t+1) = ΔS(t+1) + S(t) = -β(t)S(t)I(t) + S(t). S(t=t0) = 1 represents the number of affected road sections at the initial time of the accident. β(t) is determined according to the congestion spread coefficient of each road section in the target road network at time t, for example, it can be the average value of the congestion spread coefficient of each road section in the target road network at time t; μ(t) is determined according to the congestion dissipation coefficient of each road section in the target road network at time t, for example, it can be the average value of the congestion dissipation coefficient of each road section in the target road network at time t.

[0119] wherein, I(t) represents the number of congested road segments at time t, ΔI(t+1) represents the increment of affected road segments at time t+1, I(t+1) = ΔI(t+1) + I(t) = -β(t)S(t)I(t) - μ(t)I(t) + I(t). Wherein, S(t=t0) = 1 represents the number of affected road segments at the initial time of the accident. I(t=t0) = 1 represents the number of congested road segments at the initial time of the accident.

[0120] wherein, ΔR(t) represents the increment of recovered road segments at time t.

[0121] After determining the number of congested road segments, the congestion severity after the event can be determined based on the number of congested road segments. Wherein, the trend of the congestion severity after the event changing with time can be determined according to the ratio of the number of congested road segments at different times to the total number of road segments adjacent to the event road segment in the road network. For example, the trend of the congestion severity changing with time can be expressed as:

[0122]

[0123] wherein, N is the total number of road segments adjacent to the event road segment.

[0124] In the above embodiments, the total amount of affected road segments S(t), the number of congested road segments I(t), and the number of recovered road segments R(t) in the affected area output by the road network congestion constraint model will be input into the speed adjustment model (M4) as constraint conditions, so that the speed adjustment model predicts the speed change of each road segment at each time after the event occurs.

[0125] wherein, the speed adjustment model includes a propagation time sub-model (M4-1) for predicting the time length required for the congestion of road segment i to reach the peak value (hereinafter referred to as: event awareness time length θ e ) and the time length required for the congestion to dissipate (hereinafter referred to as: event processing time length η e ) after the event e occurs. Wherein, the event awareness time length θ e is used to reflect the perception speed of the event, i.e. the time for the congestion to spread, which can be in units of hours. The event processing time length η e is used to reflect the average time required for handling the accident, i.e. the time required for the congestion to dissipate, which can be in units of hours.

[0126] wherein, the event awareness time length θ e and the event processing time length η e of road segment i can be expressed as a function of the congestion spreading coefficient β i,t , the congestion dissipation coefficient μ i,t , the road congestion threshold ρ, and the target road network base state Net. That is:

[0127] θ i,e ,η i,e =f e (β i,t ,μ i,t Formula 4-3 (ρ,Net)

[0128]

[0129]

[0130] The target road network ground state includes the speed of different road segments in each time slice of the target road network, which can be determined based on the speed of the road segments in the historical period when they were not affected by events.

[0131] The road congestion threshold ρ is used to represent the boundary condition for judging the congestion state. A fixed value can be determined based on the definition of different congestion states on the navigation map; alternatively, the road congestion threshold can be determined using the following formula:

[0132]

[0133] in, Indicates speed during traffic congestion. This indicates the speed under free-flowing conditions on the road.

[0134] That is, the inputs to the propagation time sub-model may include: the ground state of the target road network and the congestion spread coefficient β of road segment i at time t. i,t The congestion dissipation coefficient μ of road segment i at time t i,t Event e. The output of the propagation time sub-model may include: the event perception duration θ of road segment i. e and event processing time η e .

[0135] Based on this implementation method, the propagation time sub-model can predict the event perception duration of each road segment in the target road network according to the congestion spread trend and road congestion threshold; and predict the event processing duration of each road segment in the target road network according to the congestion dissipation trend and road congestion threshold.

[0136] The speed adjustment model also includes a speed prediction sub-model (M4-2), with an event perception duration θ. e and event processing time η eThe constraint condition for speed prediction as a speed prediction sub-model, i.e., the speed prediction sub-model can predict the speed of a link within an affected area under a specific congestion spread and dissipation trend. In this embodiment, the speed prediction sub-model can predict the speed under specific constraint conditions based on a quadratic parabola model. Alternatively, the speed prediction sub-model can be expressed by the following formula:

[0137]

[0138] wherein, represents the actual speed of the projected link i at time t after the occurrence of event e; g e represents the speed adjustment function; VC j,t represents the historical speed of link j (i.e., the event link) at time t; represents the speed of link j (i.e., the event link) at time t0 (the initial time of the occurrence of the event); Δt i,j represents the time required from link i to link j; θ i,e represents the event perception time of link i after the occurrence of event e, η i,e represents the event processing time of link i after the occurrence of event e; β i,t,e represents the congestion spread coefficient of link i at time t after the occurrence of event e; V i,t represents the historical speed of link i at time t (i.e., the speed before the occurrence of event e).

[0139] In some alternative embodiments, the formula 4-5 can be expanded as the following formula:

[0140]

[0141] wherein, V i,t represents the historical speed of link i at time t (i.e., the speed at time t before the occurrence of event e); V min refers to the historical road minimum speed of link i; t0 represents the time of the start of the event. wherein, g e (β i,t,e ) represents the speed adjustment function, in some embodiments, g e (β i,t,e ) can be expressed by the following formula:

[0142] g e (β i,t,e ) = γ(M3) · max(0, 1 - β i,t,e ) formula 4-9

[0143] wherein, y() represents a conversion function from the output of the road network congestion constraint model (M3) to the speed adjustment parameter, M3 represents the output of the target road network congestion constraint model, and M3 includes S(t), I(t) and / or R(t) calculated by the foregoing embodiments.

[0144] That is, the input of the speed prediction sub-model can include: the event start time t0, the current time t, the time length Δt required from the road segment i to the road segment j i,j , the congestion spread coefficient β of the road segment i at the time t after the event e occurs i,t,e , the initial speed Vc of the event road segment (road segment j) jt0 , the speed V of the road segment i at the time t it , the event perception time length θ of the road segment i after the event e occurs ie , the event processing time length η of the road segment i after the event e occurs i,e . The output of the speed prediction sub-model can include: the actual speed V of the road segment i at the time t Further, the speed prediction sub-model can dynamically predict the speed of the road segment in the influence area under the conditions of meeting the congestion spread and dissipation trend.

[0145] Based on the foregoing embodiments, the speed prediction sub-model can predict the actual speed of each road segment in the target road network at different times after the road event occurs according to the event perception time length of each road segment in the target road network, the event processing time length of each road segment in the target road network, the speed of the event road segment at different times before and after the event occurs, and the speed of each road segment in the target road network before the road event occurs.

[0146] It should be noted that the execution subject of each step of the method provided by the foregoing embodiments can be the same device, or the method can also be executed by different devices as the execution subject. For example, the execution subject of steps 101 to 104 can be device A; for another example, the execution subject of steps 101 and 102 can be device A, and the execution subject of step 103 can be device B; and the like.

[0147] In addition, in some of the processes described in the foregoing embodiments and the accompanying drawings, a plurality of operations appearing in a certain order are included, but it should be clearly understood that these operations can be executed or executed in parallel without the order in which they appear in this text. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and these operations can be executed in sequence or in parallel. It should be noted that the “first”, “second” and the like described herein are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence. Also, “first” and “second” are not of different types.

[0148] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0149] Figure 5 The structure schematic diagram of the electronic device provided by an example embodiment of the present application is illustrated, as shown in the figure, the electronic device includes a memory 501, a processor 502 and a communication component 503. Figure 5

[0150] The memory 501 is used to store computer programs and can be configured to store other various data to support operations on the electronic device. Examples of these data include instructions for any application or method operating on the electronic device.

[0151] The processor 502 is coupled with the memory 501 and is used to execute the computer programs in the memory 501 for: determining a road event of an event road segment in a target road network; determining a detour path according to traveler information transparency and residual passing capacity of the event road segment; determining a traffic of each road segment in the target road network according to the detour path; determining a congestion spread coefficient and a congestion dissipation coefficient of each road segment in the target road network after the road event occurs according to the traffic of each road segment in the target road network; and predicting an actual speed of each road segment in the target road network after the road event occurs according to a historical speed of each road segment in the target road network under the congestion spread coefficient and the congestion dissipation coefficient of each road segment in the target road network.

[0152] Optionally, when determining the detour path according to the traveler information transparency and the residual passing capacity of the event road segment, the processor 502 is specifically used for: determining an exponential distribution model corresponding to a detour position for bypassing the event road segment; the exponential distribution model is used to determine a detour starting point and a detour arrival point from K-order neighborhood nodes of the event road segment; determining a value of a shape coefficient in the exponential distribution model according to the traveler information transparency and the residual passing capacity, so as to determine a distribution feature of the detour position in the exponential distribution model according to the value of the shape coefficient; sampling the exponential distribution model to obtain at least one detour starting point and at least one detour arrival point; and determining a detour path according to the at least one detour starting point and the at least one detour arrival point.

[0153] Optionally, the shape coefficient is positively correlated with the traveler information transparency and is negatively correlated with the residual passing capacity.​

[0154] Optionally, the processor 502 is configured to determine the detour path according to the at least one detour starting point and the at least one detour ending point, specifically by determining a plurality of candidate detour paths according to the at least one detour starting point and the at least one detour ending point; sorting the plurality of candidate detour paths according to a path quality index; the path quality index comprises a detour distance ratio and / or a path similarity; and selecting the detour path from the plurality of candidate detour paths according to a result of the sorting.

[0155] Optionally, the processor 502 is configured to determine the traffic of each road segment in the target road network according to the detour path, specifically by determining a traffic distribution ratio according to the residual traffic capacity of the event road segment; and distributing part of the traffic on the original path that needs to pass through the event road segment to the detour path according to the traffic distribution ratio, to update the traffic of each road segment in the target road network.

[0156] Optionally, the processor 502 is configured to determine the congestion spread coefficient of each road segment in the target road network after the occurrence of the road event according to the traffic of each road segment in the target road network, specifically by obtaining, for a first road segment in the target road network, traffic entering the event road segment from the first road segment as first traffic, and obtaining traffic of the event road segment as second traffic; and determining a congestion spread coefficient on the first road segment that changes over time according to a ratio of the first traffic to the second traffic and a time-varying factor; the time-varying factor is determined according to the residual traffic capacity and a time-varying traveler information transparency.

[0157] Optionally, the processor 502 is configured to determine the congestion spread coefficient of each road segment in the target road network after the occurrence of the road event according to the traffic of each road segment in the target road network, specifically by obtaining, for a first road segment in the target road network, traffic entering the event road segment from the first road segment and flowing out of the event road segment as third traffic, and obtaining traffic of the event road segment as second traffic; and determining a congestion spread coefficient on the first road segment that changes over time according to a ratio of the third traffic to the second traffic and a time-varying factor; the time-varying factor is determined according to the residual traffic capacity and a time-varying traveler information transparency.

[0158] Optionally, the processor 502 is further configured to predict the actual speed of each road segment in the target road network at different time points after the occurrence of the road event according to the historical speed of each road segment in the target road network and the congestion spread coefficient and the congestion dissipation coefficient of each road segment in the target road network, specifically configured to: predict the event perception duration of each road segment in the target road network according to the congestion spread coefficient and the road congestion threshold of each road segment in the target road network by using a speed prediction model; and predict the event processing duration of each road segment in the target road network according to the congestion dissipation coefficient and the road congestion threshold of each road segment in the target road network; and predict the actual speed of each road segment in the target road network at different time points after the occurrence of the road event according to the event perception duration of each road segment in the target road network, the event processing duration of each road segment in the target road network, the historical speed of the event road segment before the occurrence of the road event, the speed of the event road segment at different time points after the occurrence of the road event, and the speed of each road segment in the target road network before the occurrence of the road event under a preset constraint condition.

[0159] Optionally, the constraint condition includes the number of road segments affected by the road event at different time points, the number of congested road segments, and / or the number of road segments that have recovered to be unblocked; wherein the number of affected road segments, the number of congested road segments, and the number of road segments that have recovered to be unblocked at any time point are determined according to the congestion spread coefficient and the congestion dissipation coefficient of each road segment in the target road network at the time point.

[0160] Optionally, before determining the traffic of each road segment in the target road network according to the detour path, the processor 502 is further configured to: determine a first spatial range in a map according to the latitude and longitude coordinates of the road event and a specified position offset; determine a first sub-road network corresponding to the first spatial range in a road network; determine a second sub-road network in which the neighbor nodes of the specified order of the road event are located; determine the union set of the first sub-road network and the second sub-road network to obtain a target road network, and determine the road segments affected by the road event in the target road network as each road segment in the target road network.

[0161] Optionally, when determining the road segments affected by the road event in the target road network, the processor 502 is specifically configured to: for any second road segment in the target road network, determine a traffic correlation coefficient of the second road segment and the event road segment according to the traffic of the event road segment and the traffic of the event road segment flowing into the second road segment, and determine the second road segment as a road segment affected by the road event when the traffic correlation coefficient is greater than a preset first threshold; or determine a traffic change proportion of the second road segment and the event road segment according to a traffic change value of the second road segment after path detouring and the traffic of the second road segment before path detouring, and determine the second road segment as a road segment affected by the road event when the traffic change proportion is greater than a preset second threshold.

[0162] Further, as shown in Figure 5 the electronic device further includes a power supply component 504, a display component 505, an audio component 506 and other components. Figure 5 Some components are only schematically shown in the electronic device, and it does not mean that the electronic device only includes Figure 5 the components shown.

[0163] The memory 501 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0164] The communication component 503 is configured to facilitate wired or wireless communication between the device where the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as Wi-Fi, 2G (for example, Global System for Mobile Communications (GSM)), 3G (for example, Wideband Code Division Multiple Access (WCDMA)), 4G (for example, Long Term Evolution (LTE)), 4G+ (for example, LTE-Advanced (LTE-A)), or 5G (5th Generation Mobile Communication Technology), or a combination thereof. In an example embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an example embodiment, the communication component can be implemented based on a Near Field Communication (NFC) technology, a Radio Frequency IDentification (RFID) technology, an Infrared Data Association (IrDA) technology, a Ultra WideBand (UWB) technology, a Bluetooth (BT) technology, and other technologies.

[0165] The power supply component 504 is configured to supply power to various components of the device where the power supply component is located. The power supply component can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device where the power supply component is located.

[0166] The display component 505 includes a screen, which can include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide, and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touch or a slide action, but also detect a duration and a pressure associated with the touch or slide action.

[0167] The audio component 506 can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive an external audio signal when the device where the audio component is located is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory or transmitted via the communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0168] In this embodiment, after a road event occurs on an event link in a target road network, a detour path can be determined according to traveler information transparency and residual traffic capacity of the event link. The traveler information transparency is based on the perception degree of travelers to the road event, which is beneficial to more realistically simulating the travel decision of the travelers when encountering the road event. The residual traffic capacity of the event link is used as a prediction basis to improve the accuracy of the prediction result. According to the detour path, the traffic of each link in the target road network can be determined, and according to the traffic of each link in the target road network, the congestion spreading trend and the congestion dissipation trend of each link in the target road network after the road event occurs can be determined. The congestion spreading trend and the congestion dissipation trend are determined based on the relatively reliable prediction result of the traveler decision, and thus can more truly reflect the influence of the target road network by the road event. Further, under the congestion spreading trend and the congestion dissipation trend of each link in the target road network, the actual speed of each link in the target road network after the road event occurs, which is predicted according to the historical speed of each link in the target road network, can more accurately reflect the traffic state of the target road network under the influence of the road event.

[0169] Correspondingly, the embodiment of the present application further provides a computer readable storage medium storing a computer program, and the computer program can implement each step which can be executed by the electronic device when the computer program is executed. The computer program can be realized as an application program running in the electronic device.

[0170] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of being implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0171] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0172] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0173] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0174] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0175] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory, etc. The memory is an example of computer readable media.

[0176] Computer readable media includes permanent and non-permanent, moveable and non- moveable media that can be implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disks (DVDs) or other optical storage, magnetic cassettes, magnetic disks storage or other magnetic storage devices, or any other non-transmission medium that is used to store information that is accessible to a computing device. According to the definition provided herein, computer readable media excludes transitory media, such as modulated data signals and carrier waves.

[0177] It should also be noted that the terms "comprising," "including," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0178] The above description is merely illustrative of the application, and not restrictive. Various modifications and changes can become apparent to those skilled in the art. Incorporating any modification, equivalent substitution, improvement, etc. within the spirit and principle of the application, shall be included in the scope of the claims of the application.

Claims

1. A speed prediction method characterized by, The method comprises: determining a road event of an event road section in a target road network; determining a detour path according to an information transparency of travelers and a residual passing capacity of the event road section, wherein the information transparency of the travelers is used to describe that a path decision rule of the travelers is affected by traffic situation information currently mastered by the travelers after the road event occurs; determining a traffic flow of each road section in the target road network according to the detour path; determining a congestion spreading coefficient and a congestion dissipation coefficient of each road section in the target road network after the road event occurs according to the traffic flow of each road section in the target road network; predicting an actual speed of each road section in the target road network after the road event occurs according to a historical speed of each road section in the target road network under the congestion spreading coefficient and the congestion dissipation coefficient of each road section in the target road network.

2. The method of claim 1, wherein, The method of determining a detour path according to an information transparency of travelers and a residual passing capacity of an event road section comprises: determining an exponential distribution model corresponding to a detour position for bypassing the event road section, wherein the exponential distribution model is used to determine a detour starting point and a detour arrival point from K-order neighborhood nodes of the event road section; determining a value of a shape coefficient in the exponential distribution model according to the information transparency of the travelers and the residual passing capacity, so as to determine a distribution feature of the detour position in the exponential distribution model according to the value of the shape coefficient; sampling the exponential distribution model to obtain at least one detour starting point and at least one detour arrival point; determining a detour path according to the at least one detour starting point and the at least one detour arrival point.

3. The method of claim 2, wherein, The shape coefficient is positively correlated with the information transparency of the travelers and is negatively correlated with the residual passing capacity.

4. The method of claim 2, wherein, The method of determining a detour path according to the at least one detour starting point and the at least one detour arrival point comprises: determining a plurality of candidate detour paths according to the at least one detour starting point and the at least one detour arrival point; sorting the plurality of candidate detour paths according to a path quality index, wherein the path quality index at least comprises a detour distance ratio and / or a path similarity; screening a detour path from the plurality of candidate detour paths according to a result of the sorting.

5. The method of claim 1, wherein, The method of determining a traffic flow of each road section in a target road network according to a detour path comprises: determining a traffic flow distribution ratio according to a residual passing capacity of an event road section; distributing part of a traffic flow on an original path passing through the event road section to the detour path according to the traffic flow distribution ratio, so as to update the traffic flow of each road section in the target road network.

6. The method of claim 1, wherein, The method of determining a congestion spreading coefficient of each road section in a target road network after a road event occurs comprises: for a first road section in the target road network, obtaining a traffic flow entering the event road section from the first road section as a first traffic flow, and obtaining a traffic flow of the event road section as a second traffic flow; determine a congestion spread coefficient of the first road segment over time according to a ratio of the third flow and the second flow and a time-varying factor, wherein the time-varying factor is determined according to the residual capacity and a time-varying information transparency of travelers.

7. The method of claim 1, wherein, determine a congestion dissipation coefficient of each road segment in the target road network after the road event according to flows of the road segments in the target road network, including: for a first road segment in the target road network, obtain a flow entering the event road segment from the first road segment and flowing out of the event road segment as a third flow, and obtain a flow of the event road segment as a second flow; determine a congestion dissipation coefficient of the first road segment over time according to a ratio of the third flow and the second flow and a time-varying factor, wherein the time-varying factor is determined according to the residual capacity and a time-varying information transparency of travelers.

8. The method of claim 1, wherein, predict actual speeds of the road segments in the target road network after the road event according to historical speeds of the road segments in the target road network under the congestion spread coefficients and the congestion dissipation coefficients of the road segments in the target road network, including: use a speed prediction model to predict an event perception duration of each road segment in the target road network according to the congestion spread coefficient of the road segment and a road congestion threshold, and predict an event processing duration of each road segment in the target road network according to the congestion dissipation coefficient of the road segment and the road congestion threshold; predict actual speeds of the road segments in the target road network at different time instants after the road event according to the event perception duration of each road segment in the target road network, the event processing duration of each road segment in the target road network, a historical speed of the event road segment before the road event, a speed of the event road segment at different time instants after the road event, and historical speeds of the road segments in the target road network before the road event, under a set constraint condition.

9. The method of claim 8, wherein, the constraint condition includes a number of affected road segments, a number of congestion road segments, and / or a number of recovered road segments at different time instants affected by the road event, wherein the number of affected road segments, the number of congestion road segments, and the number of recovered road segments at any time instant are determined according to the congestion spread coefficient and the congestion dissipation coefficient of each road segment in the target road network at the time instant.

10. The method according to any one of claims 1 to 9, characterized in that, before determining the flows of the road segments in the target road network according to the detour path, further including: determine a first spatial range in a map according to the latitude and longitude coordinates of the road event and a specified position offset; determine a first sub-road network corresponding to the first spatial range in a road network, and determine a second sub-road network in which a specified order of neighborhood nodes of the road event are located; determine a union set of the first sub-road network and the second sub-road network to obtain a target road network, and determine road segments affected by the road event in the target road network as the road segments in the target road network.

11. The method of claim 10, wherein, determine road segments affected by the road event in the target road network as the road segments in the target road network, including: For any second road segment in the target road network, a traffic correlation coefficient between the second road segment and the event road segment is determined according to the traffic of the event road segment and the traffic of the event road segment flowing into the second road segment, and when the traffic correlation coefficient is greater than a preset first threshold, the second road segment is determined as a road segment affected by the road event; or, According to the traffic change value of the second road segment after path detouring and the traffic of the second road segment before path detouring, a traffic change proportion of the second road segment and the event road segment is determined, and when the traffic change proportion is greater than a preset second threshold, the second road segment is determined as a road segment affected by the road event.

12. An electronic device, comprising: Comprise: a memory and a processor; the memory is used to store one or more computer instructions; the processor is used to execute the one or more computer instructions for: executing the steps in the method of any one of claims 1-11.

13. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to realize the speed prediction method of any one of claims 1-11.

Images