Trajectory generation method, device and storage medium

Abstract

Embodiments of the present application provide a trajectory generation method and device and a storage medium, wherein the method comprises: obtaining at least one group of node road segments contained in a target region in a target period; performing path planning on the at least one group of node road segments respectively to obtain trajectories generated by the at least one group of node road segments respectively; and generating a trajectory of the target region in the target period according to the trajectories generated by the at least one group of node road segments respectively. Accordingly, the comprehensiveness and / or accuracy of the trajectory of the target region in the target period generated in the embodiments of the present application can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a trajectory generation method, device and storage medium. BACKGROUND

[0002] In many scenarios such as road traffic management, a large number of processing links need to rely on trajectories in a target area.

[0003] At present, trajectories of moving objects in a target area are mainly collected by relying on a carrier such as a navigation software which can record travel information of the moving objects, but the trajectories obtained by this way are one-sided, which will seriously affect the accuracy of processing links based on the trajectories. SUMMARY

[0004] Aspects of the present application provide a trajectory generation method, device and storage medium to improve comprehensiveness and / or accuracy of trajectories.

[0005] An embodiment of the present application provides a trajectory generation method, comprising:

[0006] obtaining at least one group of node road segments contained in a target area in a target period;

[0007] respectively performing path planning on the at least one group of node road segments to obtain trajectories respectively generated by the at least one group of node road segments;

[0008] generating a trajectory of the target area in the target period according to the trajectories respectively generated by the at least one group of node road segments.

[0009] An embodiment of the present application further provides a computing device comprising a memory and a processor.

[0010] The memory is configured to store one or more computer instructions.

[0011] The processor is coupled with the memory and is configured to execute the one or more computer instructions to:

[0012] obtain at least one group of node road segments contained in a target area in a target period;

[0013] respectively perform path planning on the at least one group of node road segments to obtain trajectories respectively generated by the at least one group of node road segments;

[0014] generate a trajectory of the target area in the target period according to the trajectories respectively generated by the at least one group of node road segments.

[0015] The embodiment of the present application further provides a computer readable storage medium storing computer instructions, which, when executed by one or more processors, cause the one or more processors to perform the trajectory generation method.

[0016] In the embodiment of the present application, the abstract features of the trajectory contained by the target region in the target period are characterized by at least one set of node link, and by comprehensively mining the abstract features, the trajectory contained by the target region in the target period can be classified, so that the trajectory contained by the target region in the target period can be more comprehensively summarized to improve the comprehensiveness of the generated trajectory; the trajectory reduction processing of the target region in line with the path planning experience under the guidance of the abstract features can generate the trajectory contained by the target region in the target period. This makes the generated trajectory meet the abstract features of the trajectory contained by the target region in the target period and the path planning experience, and is closer to the real trajectory contained by the target region in the target period. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, but do not constitute improper limitations on the present application. In the drawings:

[0018] Figure 1 A flowchart of a trajectory generation method provided by an embodiment of the present application is shown in the figure;

[0019] Figure 2 A schematic diagram of an application scenario provided by an embodiment of the present application is shown in the figure;

[0020] Figure 3 A structural schematic diagram of a computing device provided by another embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0021] To make the purpose, technical scheme and advantages of the present application clearer, the technical scheme of the present application will be described clearly and completely below by combining the specific embodiments of the present application with 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 labor fall within the scope of protection of the present application.

[0022] Currently, the trajectory generated for the target area is one-sided and has low accuracy. To solve these problems, in some embodiments of the present application: at least one set of node road segments is used to represent the abstract features of the trajectory contained in the target area in the target period, and by comprehensively mining these abstract features, the trajectory contained in the target area in the target period can be classified, so that the trajectory contained in the target area in the target period can be more comprehensively summarized to improve the comprehensiveness of the generated trajectory; and the trajectory reduction processing of the target area in line with the path planning experience under the guidance of the abstract features can generate the trajectory contained in the target area in the target period. This makes the generated trajectory meet the abstract features of the trajectory contained in the target area in the target period and the path planning experience, and is closer to the real trajectory contained in the target area in the target period.

[0023] The technical solutions provided by the embodiments of the present application will be described in detail below with reference to the drawings.

[0024] Figure 1 A flowchart of a trajectory generation method provided by an embodiment of the present application is shown. The trajectory generation method provided by the embodiment can be executed by a trajectory generation device, which can be implemented as software or as a combination of software and hardware, and can be integrated into a computing device. As shown in the figure, the method comprises: Figure 1

[0025] Step 100: obtaining at least one set of node road segments contained in the target area in the target period;

[0026] Step 101: performing path planning on the at least one set of node road segments respectively to obtain the trajectories generated by the at least one set of node road segments respectively;

[0027] Step 102: generating the trajectory contained in the target area in the target period according to the trajectories generated by the at least one set of node road segments respectively.

[0028] The trajectory generation method provided by the embodiment can be applied in various application scenarios that require trajectory generation. For example, it can be applied in road traffic application scenarios such as road condition evaluation, travel analysis, traffic optimization or traffic simulation. For another example, it can also be applied in waterway, airway or logistics application scenarios. For another example, it can also be applied in transportation hubs and surrounding application scenarios such as subway stations, light rail stations, high-speed rail stations, airports, bus stations, ports, logistics stations, etc. The application does not limit the application scenarios.

[0029] In addition, the trajectory generation method provided by the embodiment can be used in scenarios of restoring trajectories in the past period, and can also be used in scenarios of predicting trajectories in the future period.

[0030] ​The specification of the target area can be set according to actual needs. The specification of the target area can be a district, a county, a city, and the like, and of course, can also be other customized specifications, which are not limited in the embodiment.

[0031] The target area includes a plurality of node road segments. The node road segments can be popular road segments or random road segments in the target area, and of course, can also be road segments selected from the target area according to use requirements. In actual application, a road segment in the target area on which a traffic sensing device is deployed can be used as a node road segment. The traffic sensing device can be a camera or a ground induction coil and the like, which can sense the traffic of a moving object.

[0032] The target period can be a time slice in the target cycle. For example, when the target cycle is set as a natural day, the target period can be a time slice in a natural day, for example, the target period can be 10:00-10:05 in the morning. Of course, the length of the target period is not limited in the embodiment.

[0033] In the embodiment, an interactive interface can be provided, and a user can configure cycle, period, area and the like task parameters in the interactive interface to create a trajectory generation task. The trajectory generation task can trigger execution of each step in the embodiment. Of course, execution of each step in the embodiment can also be triggered based on other events, for example, a trajectory acquisition request sent by other equipment can be received, thereby triggering execution of each step in the embodiment. Of course, the embodiment is not limited thereto.

[0034] In step 100, at least one group of node road segments included by the target area in the target period can be acquired. The at least one group of node road segments is used to represent the abstract features of the trajectories included by the target area in the target period.

[0035] In actual application, in addition to the node road segments, the at least one group of node road segments can also include attribute information of the node road segments. The attribute information of the node road segments includes the order between the node road segments, the time distribution of entering each node road segment, and the like. Accordingly, a group of node road segments can be represented as L C ={X, T}, wherein X={A, B, C, D…}, T={t A ,t B ,t C ,t D …}; A, B, C, D… represent node road segments, and t A ,t B ,t C ,t D … represent the time distribution of entering each node road segment.

[0036] In the embodiment, a plurality of ways can be used to obtain at least one group of node road segments included by the target area in the target period.

[0037] In an implementation, a plurality of sample trajectories in the target region and adaptive to the target period can be acquired; the plurality of sample trajectories can be clustered based on a mapping relationship between the sample trajectories and the node road segments, to obtain at least one type of sample trajectories; and the node road segments and attribute information of the node road segments contained in the at least one type of sample trajectories can be extracted respectively, to generate a group of node road segments corresponding to each of the at least one type of sample trajectories.

[0038] The sample trajectories can be obtained from various navigation data or artificial statistical data, and the like. Taking the navigation data as an example, in actual application, communication can be established with a navigation data provider, a navigation data acquisition request can be initiated to the navigation data provider, and the navigation data acquisition request can contain information of the target period and the target region, so as to enable the navigation data provider to find a plurality of sample trajectories in the target region and adaptive to the target period and to feed back the sample trajectories. Accordingly, the sample trajectories can be acquired.

[0039] In the implementation, a plurality of sample trajectories matching the target period can be selected from sample trajectories occurring in the target region. The sample trajectories matching the target period refer to sample trajectories whose first node road segment occurs in the target period.

[0040] Based on the road network structure of the target region, a mapping relationship between the sample trajectories and the node road segments can be established, and the mapping relationship can be used to represent the node road segments set on the sample trajectories. Accordingly, the plurality of sample trajectories can be clustered according to the node road segments contained in each of the plurality of sample trajectories, to obtain at least one type of sample trajectories.

[0041] In actual application, the node road segments and / or the attribute information of the node road segments contained in different types of sample trajectories are different. Of course, the embodiments are not limited thereto, and different types of sample trajectories can also be distinguished from other dimensions.

[0042] On this basis, the node road segments and the attribute information of the node road segments contained in the at least one type of sample trajectories can be extracted respectively, to generate a group of node road segments corresponding to each of the at least one type of sample trajectories. That is, a type of sample trajectories can be represented by a group of node road segments.

[0043] In the implementation, at least one group of node road segments contained in the target region in the target period can be extracted based on a sufficient number of sample trajectories, to comprehensively estimate the abstract features of various trajectories contained in the target region in the target period, so that the trajectories contained in the target region in the target period can be summarized more accurately, and the comprehensiveness of the generated trajectories can be improved.

[0044] It is worth noting that the sample trajectories mentioned herein are composed of continuous road segments, and the sample trajectories describe the road segments contained therein and the time information of entering each road segment.

[0045] In another implementation, the sample trajectory in the former implementation can be replaced by a trajectory estimated based on a path selection algorithm.

[0046] In practice, a location hotspot contained in the target region in the target period can be acquired, and a path selection algorithm is performed based on the location hotspot to generate a plurality of reference trajectories. The reference trajectories are taken as sample trajectories to generate at least one group of node road segments contained in the target region in the target period. The location hotspot can be taken as a starting point or an ending point of the reference trajectory, and the path selection algorithm can employ an N-shortest path algorithm.

[0047] Of course, the above two implementations are only exemplary, and in the present embodiment, other implementations can also be employed to acquire at least one group of node road segments contained in the target region in the target period, and the present embodiment is not limited thereto.

[0048] On this basis, in step 101, path planning can be performed on the at least one group of node road segments to obtain trajectories generated by the at least one group of node road segments respectively.

[0049] For each group of node road segments, path planning can be performed to determine a trajectory generated by the group of node road segments. The trajectory generated by the group of node road segments all conforms to the abstract features represented by the group of node road segments. In practice, all the node road segments in the group of node road segments are contained in the trajectory generated by the group of node road segments, and the trajectory conforms to the attribute information of the node road segments.

[0050] In this way, based on the at least one group of node road segments, a plurality of trajectories conforming to the abstract features represented by the at least one group of node road segments can be derived.

[0051] In step 102, based on these trajectories, a trajectory contained in the target region in the target period can be generated. The trajectory contained in the target region in the target period refers to a trajectory generated by a mobile object in the target region moving in the target period. The mobile object can be a motor vehicle, a person, a logistics vehicle, a ship, an aircraft, an electric vehicle, a bicycle, or other types of objects, and the present embodiment is not limited thereto.

[0052] Accordingly, the embodiments of the present application have at least the following beneficial effects:

[0053] 1. The abstract features of the trajectory contained in the target region in the target period are represented by the at least one group of node road segments, and by comprehensively mining these abstract features, classification of the trajectory contained in the target region in the target period can be achieved, so that the trajectory contained in the target region in the target period can be more comprehensively summarized, to improve the comprehensiveness of the generated trajectory.

[0054] 2. The path planning is performed on the at least one group of node road segments to derive a plurality of trajectories conforming to the abstract features represented by the at least one group of node road segments, and then the trajectories are used to generate trajectories included in the target area in the target period, which ensures that the generated trajectories conform to the path planning experience, thereby improving the accuracy of the generated trajectories.

[0055] 3. The target area and the target period can be flexibly set as needed, thereby improving the application scenario range of the scheme and meeting the requirements of different application scenarios in terms of scale and / or time granularity.

[0056] 4. The trajectory restoration processing is performed on the target area under the guidance of the at least one group of node road segments, which can reduce the dependence of the trajectory generation process on real user data, thereby protecting the privacy of real user data.

[0057] In the above or the following embodiments, a location hotspot associated with each of the at least one group of node road segments can be obtained, and the location hotspot is a location point outside the at least one group of node road segments in the target area; the trajectories generated by the at least one group of node road segments are extended according to the location hotspots associated with the at least one group of node road segments, to obtain the trajectories included in the target area in the target period.

[0058] The target area in the target period can include a plurality of location hotspots. The plurality of location hotspots can be determined by counting the start points and end points of sample trajectories occurring in the target area. For example, a location point whose number or probability of being a start point or an end point of a sample trajectory exceeds a preset standard can be regarded as a location hotspot. Accordingly, for a location hotspot, it can be a start point or an end point of a trajectory included in the target area in the target period.

[0059] As mentioned above, the at least one group of node road segments is a clustering result of sample trajectories. Therefore, in actual applications, the start points and end points of the sample trajectories summarized by the at least one group of node road segments can be regarded as the location hotspots associated with the at least one group of node road segments.

[0060] On this basis, the trajectories generated by the at least one group of node road segments can be extended to the location hotspots associated therewith to form complete trajectories with the location hotspots as start points or end points, and these trajectories can be regarded as the trajectories included in the target area in the target period.

[0061] That is, in this embodiment, the trajectories generated by performing path planning on the at least one group of node road segments are extended to the location hotspots in the target area in the target period, thereby generating trajectories conforming to real start points and real end points, which makes the generated trajectories closer to real trajectories.

[0062] In this embodiment, the extension of the trajectories generated by each of the at least one group of node road segments can be implemented in various ways. Since the extension schemes of the trajectories generated by each group of node road segments are similar, for ease of description, the following will take the first group of node road segments as an example to describe the schemes, and it should be understood that the first group of node road segments can be any one of the at least one group of node road segments.

[0063] In an implementation, for the first trajectory generated by the first group of node road segments, a target location hotspot can be selected for the first trajectory from the location hotspots associated with the first group of node road segments; based on the target location hotspot, the first trajectory can be extended to obtain an accessible trajectory between the first trajectory and the target location hotspot; and the first trajectory and the accessible trajectory can be connected to obtain a trajectory of the target region in the target period. The first trajectory is any one of the trajectories generated by the first group of node road segments.

[0064] In this implementation, the location hotspots associated with the first group of node road segments can be sampled to select a target location hotspot for the first trajectory. The sampling method can be random sampling or sampling according to the heat, etc., which is not limited here. Generally, the target location hotspot selected for the first trajectory is a pair, i.e., two target location hotspots respectively serving as the starting point and the ending point, but this is not limited here. In actual application, the target location hotspot selected for the first trajectory can be one, and can also be zero, because the sample trajectories summarized by the first group of node road segments can include sample trajectories with starting points and / or ending points located on the node road segments.

[0065] After determining the target location hotspot corresponding to the first trajectory, the path planning can be performed with the target location hotspot and the end node road segment of the first trajectory as the starting point and the ending point, respectively, to determine the accessible trajectory between the first trajectory and the target location hotspot.

[0066] Based on this, the first trajectory and the accessible trajectory can be connected to form a trajectory, which will serve as a trajectory of the target region in the target period.

[0067] In another implementation, each trajectory generated by the first group of node trajectories can be extended to the location hotspots associated with the first group of node road segments to obtain several accessible trajectories between each trajectory generated by the first group of node trajectories and the location hotspots associated with the first group of node road segments; sampling of the accessible trajectories can be performed for each trajectory generated by the first group of node trajectories to determine the corresponding accessible trajectory for each trajectory generated by the first group of node trajectories; and each trajectory generated by the first group of node trajectories and the corresponding accessible trajectory can be connected to generate partial trajectories of the target region in the target period.

[0068] Wherein, the reachable trajectory can be generated by path planning. In addition, the sampling method of the reachable trajectory can be random sampling or sampling according to the heat of the location hotspot, which is not limited herein.

[0069] Compared with the previous implementation mode, the implementation mode adjusts the order of the generation stage and the sampling stage of the reachable trajectory.

[0070] Of course, the above implementation mode is only exemplary, and the present embodiment is not limited thereto.

[0071] In the present embodiment, the trajectory generated based on at least one set of node road segments can be extended to the location hotspot of the target region in the target period, and the trajectory generated after the extension will be closer to the real trajectory of the target region in the target period.

[0072] In the above or the following embodiments, the number of trajectories corresponding to each of the at least one set of node road segments can be obtained; and path planning is performed on the at least one set of node road segments respectively to obtain the trajectories generated by each of the at least one set of node road segments and meeting the number of trajectories corresponding thereto.

[0073] As mentioned above, based on at least one set of node road segments, the classification and summarization of the trajectories contained in the target region in the target period can be realized, and in the present embodiment, a sufficient number of trajectories can be generated under each trajectory category to generate the full amount of trajectories contained in the target region in the target period.

[0074] Wherein, the full amount of trajectories refers to all the trajectories generated by the moving objects in the target region in the target period. It is worth noting that the full amount of trajectories in the present embodiment is a relative concept, that is, all the trajectories relative to the preset range. The preset range in the present embodiment refers to the number of trajectories corresponding to each of the at least one set of node road segments.

[0075] In the present embodiment, various implementation modes can be used to obtain the number of trajectories corresponding to each of the at least one set of node road segments.

[0076] In one implementation mode, the number of trajectories corresponding to each of the at least one set of node road segments can come from an empirical value. For example, it can come from the statistical results of navigation software or from the estimated value provided by experts, etc. Especially, when the trajectory generation method provided by the present embodiment is used to predict the trajectories in the future period, the implementation mode can be used to determine the number of trajectories corresponding to each of the at least one set of node road segments.

[0077] In another implementation, the number of sample trajectories included in at least one type of sample trajectory and the number of periods can be determined according to the aforementioned embodiments, and the number of sample trajectories included in at least one type of sample trajectory in a single period can be determined; and the ratio of the number of sample trajectories included in at least one type of sample trajectory in a single period to the sample trajectory penetration rate corresponding to the target area can be calculated as the respective trajectory number of at least one group of node road segments.

[0078] In this implementation, the sample trajectories of one type corresponding to a group of node road segments can come from multiple periods, and the number of sample trajectories included in one type of sample trajectory can be converted into the number of sample trajectories in the target period in a single period as the reference trajectory number of the corresponding group of node road segments according to the number of periods.

[0079] For example, in the case of using 30-day sample trajectories in the target area, if the period is one day and the target period is 10:00-10:05, the sample trajectories included in one type of sample trajectory come from 10:00-10:05 of each day in 30 days, and based on this, the number of sample trajectories included in one type of sample trajectory can be divided by 30 to obtain the number of sample trajectories in 10:00-10:05 in one day (i.e., the number of trajectories in the target period in a single period), which is used as the reference trajectory number of the corresponding group of node road segments corresponding to this type of sample trajectory. Of course, this embodiment is not limited to this way.

[0080] In this implementation, the number of sample trajectories included in at least one type of sample trajectory in a single period and the sample trajectory penetration rate corresponding to the target area can be used to calculate the respective trajectory number of at least one group of node road segments. This is especially suitable for the case where the sample trajectories are sufficient.

[0081] In another implementation, the mobile object flow of each node road segment in the target area in the target period can be obtained; and the respective trajectory number of at least one group of node road segments that meets the target is calculated, with the difference between the estimated flow of each node road segment and the mobile object flow collected by each node road segment being minimized; wherein the estimated flow is determined according to the respective trajectory number of at least one group of node road segments and the node road segments included in each group.

[0082] In this implementation, the flow sensing device on each node road segment can be used to collect the mobile object flow of each node road segment in the target period. For example, the aforementioned camera or ground inductive coil.

[0083] In actual application, a sparse matrix A can be used to represent the node road segments included in at least one group of node road segments in the target period of the target area. A = [a jl, where j = 0, 1, 2…J-1, J represents the number of node links in the target area; and l = 0, 1, 2…L-1, L represents the number of groups of node links contained in the target area in the target period.

[0084] For example, the sparse matrix

[0085] The rows in the matrix A represent 5 node links contained in the target area, the columns in the matrix A represent 4 groups of node links contained in the target area in the target period, and the element of 1 in the matrix indicates that the group of node links corresponding to the row contains the node link corresponding to the column.

[0086] Accordingly, the vector is introduced to represent the number of trajectories corresponding to each group of node links, and the vector is introduced to represent the mobile object flow collected on each node link, and the first expression is constructed: wherein is the object to be solved, and U = diag{min{u j -2 , δ}}. j is the standard deviation of f j , and δ is an arbitrarily small positive number. Then, the CVXOPT or other solvers can be used to solve the quadratic programming problem corresponding to the expression, and the

[0087] In this implementation, the number of trajectories corresponding to each group of node links is determined based on the mobile object flow collected on each node link of the target area in the target period, so that the number of trajectories corresponding to each group of node links can match the mobile object flow collected on each node link of the target area in the target period. Especially when the flow sensing device is arranged on the node link, the number of trajectories corresponding to each group of node links can match the real flow observation value in the target area, so that the generated trajectories are closer to the real full-quantity trajectories in the target area in terms of quantity.

[0088] In yet another implementation, the first two implementations can be combined: determining the number of sample trajectories included in at least one type of sample trajectory in a single period according to the number of sample trajectories and the number of periods included in each of the at least one type of sample trajectory; calculating the ratio of the number of sample trajectories included in at least one type of sample trajectory in a single period to the sample trajectory penetration rate corresponding to the target area, as the reference trajectory number corresponding to each of the at least one group of node road segments, respectively; obtaining the mobile object flow collected by each node road segment in the target period; and taking the sum of the difference between the estimated flow of each node road segment and the mobile object flow collected by the node road segment and the difference between the trajectory number and the reference trajectory number corresponding to each of the at least one group of node road segments as the target, calculating the trajectory number corresponding to each of the at least one group of node road segments; wherein the estimated flow is determined according to the trajectory number corresponding to each of the at least one group of node road segments and the node road segments included in each of the at least one group of node road segments.

[0089] The trajectory number corresponding to each of the at least one group of node road segments obtained in this way not only refers to the experience information provided by the sample trajectories, but also refers to the real flow observation value on the node road segments, which makes the trajectory number corresponding to each of the at least one group of node road segments more accurate, more reasonable, and closer to reality.

[0090] In the foregoing implementations, the sample trajectory penetration rate of the target area is mentioned, which refers to the proportion of the number of sample trajectories in the target area to the total number of trajectories. The sample trajectory penetration rate is usually provided by the provider of the sample trajectories. In some cases, the sample trajectory penetration rate of the target area can also be calculated according to the mobile object flow collected on the node road segments and the situation of the node road segments passed through by the sample trajectories.

[0091] In practical applications, the second expression can be constructed: wherein f j is the mobile object flow collected by the jth node road segment, is the flow of the sample trajectories passing through the jth node road segment, and δ is an arbitrary small positive number.

[0092] Accordingly, the trajectory number corresponding to each of the at least one group of node road segments can be obtained.

[0093] On this basis, path planning can be performed on each of the at least one group of node road segments according to the trajectory number corresponding to each of the at least one group of node road segments, respectively, to obtain trajectories generated by each of the at least one group of node road segments in accordance with the trajectory number.

[0094] For example, if the trajectory number corresponding to a group of node road segments is 10, then 10 trajectories are generated for the group of node road segments. The 10 trajectories generated will be extended to the location hotspots in the trajectory extension link in the foregoing embodiments, thereby generating 10 trajectories included in the target area in the target period.

[0095] In the above or the following embodiments, path planning can be performed on at least one group of node road segments. Since the path planning process for each group of node road segments is similar, for ease of description, in this embodiment, the path planning process will be described taking the first group of node road segments as an example. It should be understood that the first group of node road segments can be any one of the at least one group of node road segments.

[0096] In this embodiment, path planning can be performed on the first group of node road segments to determine the connection trajectories between each two adjacent node road segments in the first group of node road segments. The number of trajectories corresponding to the first group of node road segments is used as the sampling times, and the connection road segments between each two adjacent node road segments in the first group of node road segments are sampled. Each sampling result includes a target connection trajectory between each two adjacent node road segments in the first group of node road segments. The target connection trajectory in each sampling result is connected to the node road segments in the first group of node road segments, respectively, to obtain trajectories generated by the first group of node road segments that conform to the number of trajectories thereof.

[0097] Here, adjacent node road segments refer to two node road segments that are adjacent in sequence, and the adjacency here is independent of geographical position. In actual applications, the node road segments included in the first group of node road segments can be discrete. In this embodiment, path planning can be performed between each two adjacent node road segments in the first group of node road segments to determine the connection trajectories between each two adjacent node road segments.

[0098] In this way, a plurality of connection trajectories will be generated between each two adjacent node road segments in the first group of node road segments. On this basis, the number of trajectories corresponding to the first group of node road segments can be used as the sampling times, and the connection road segments between each two adjacent node road segments in the first group of node road segments can be sampled. For example, if it is determined that the number of trajectories corresponding to the first group of node road segments is 10, the connection road segments between each two adjacent node road segments in the first group of node road segments can be sampled 10 times. Each time sampling is performed on the connection paths between each two adjacent node road segments in the first group of node road segments to obtain a target connection path between each two adjacent node road segments, and accordingly 10 sampling results can be obtained.

[0099] Subsequently, the 10 sampling results are connected to the node road segments included in the first group of node road segments, respectively, to obtain 10 trajectories corresponding to the first group of node road segments. In actual applications, connection refers to connecting the target connection trajectories to the discrete node road segments according to the sequence of the node road segments included in the first group of node road segments and the association between the target connection trajectories and the node road segments, to form a continuous trajectory.

[0100] The generation scheme of the connection track between two adjacent node road sections will be described in detail below. For ease of description, the following will take the adjacent first node road section and the second node road section in the first group of node road sections as an example. The first node road section and the second node road section can be any two adjacent node road sections in the first group of node road sections.

[0101] In this embodiment, for the adjacent first node road section and the second node road section, the probability of the current road section entering the second node road section through its adjacent downstream road section can be calculated starting from the first node road section; based on the probability, the next road section of the current road section is determined until the second node road section is reached, so as to generate the connection track between the first node road section and the second node road section.

[0102] After the first node road section, it can take several road sections to reach the second node road section. For the current road section, when it is determined based on the road network structure that it has one or more downstream road sections, the probability of the current road section entering the second node road section through each of its adjacent downstream road sections can be calculated in this embodiment.

[0103] For example, if the current road section rid1 has three downstream road sections rid2a, rid2b and rid2c, the probability of rid1 entering the second node road section through rid2a, the probability of rid1 entering the second node road section through rid2b, and the probability of rid1 entering the second node road section through rid2c can be calculated respectively.

[0104] Accordingly, in this embodiment, starting from the first node road section, the selection of the subsequent road section can be gradually performed according to the above-mentioned probability until the second node road section is reached, so as to generate multiple connection tracks between the first node road section and the second node road section.

[0105] In order to calculate the above-mentioned probability, in this embodiment, the first probability of the current road section entering the second node road section through its adjacent downstream road section can be determined according to the sample track in the target area; the second probability of the current road section entering the second node road section through its adjacent downstream road section can be determined based on the shortest path principle; the sample penetration rate corresponding to the target area is taken as the weight of the first probability, and the difference obtained by subtracting the sample penetration rate from 1 is taken as the weight of the second probability, and the first probability and the second probability are weighted and summed to determine the probability of the current road section entering the second node road section through its adjacent downstream road section.

[0106] That is, for each downstream road section adjacent to the current road section, the one-dimensional probability of the current road section entering the second node road section through each of its adjacent downstream road sections can be estimated from the sample track dimension and the shortest path principle dimension respectively; and the probabilities estimated from the two dimensions are fused to finally determine the probability of the current road section entering the second node road section through each of its adjacent downstream road sections.

[0107] P1(i→j) represents the single-dimension probability of the current road segment rid i adjacent downstream road segment rid j P2(i→j) represents the single-dimension probability of the current road segment rid i adjacent downstream road segment rid j adjacent downstream road segment rid i adjacent downstream road segment rid j adjacent downstream road segment rid i wherein, a represents the sample penetration rate of the target area.

[0108] The following will take the first downstream road segment adjacent to the current road segment as an example to describe the process of estimating the single-dimension probability of the current road segment rid

[0109] From the dimension of sample trajectories, in the embodiment, for the first downstream road segment adjacent to the current road segment, based on the set of sample trajectories in the target area that match the attribute information of the first group of node road segments, the proportion of the sample trajectories from the current road segment to the second node road segment through the first downstream road segment to the sample trajectories from the current road segment to the second node road segment is counted, and according to the proportion, the first probability corresponding to the first downstream road segment is determined.

[0110] In the embodiment, in order to improve the accuracy of the first probability, the set of sample trajectories that match the attribute information of the first group of node road segments is selected from the sample trajectories in the target area, that is, in the embodiment, the sample trajectories used to calculate the first probability need to meet the requirements of the order between the node road segments in the first group of node road segments and the time distribution of entering each node road segment, so that the calculated first probability meets the road conditions of the target area in the target period. After all, the road conditions in different periods may be different, and the results of path planning will also be different.

[0111] In the embodiment, the ngram model can be referred to to count the first probability of the current road segment rid wherein, C(rid i , rid j ) represents the number of sample trajectories from the current road segment rid i to the second node road segment through the first downstream road segment rid j . i) represents the first probability of the current road segment rid i the number of sample trajectories entering the second node road segment.

[0112] In practical applications, the first probability can be rewritten as: P1(i→j) = P(rid j #t j |rid i #t i , D, t0), that is, the number of sample trajectories entering the first node road segment at time t0is counted, and the time t i entering the current road segment rid i and going to the destination road segment D (i.e., the second node road segment), the probability of selecting the downstream road segment rid j , and the time t j entering the downstream road segment rid j . Among them, the time t i entering the current road segment rid i can be determined when the probability of the previous road segment to the current road segment is estimated.

[0113] Accordingly, the first probability of the current road segment entering the second node road segment through the first downstream road segment can be determined, and the time of entering the first downstream road segment in the connected trajectory can be determined as a parameter of the connected trajectory.

[0114] From the dimension of the shortest path principle, in this embodiment, for the first downstream road segment adjacent to the current road segment, the distance from the current road segment to the second node road segment through the first downstream road segment is calculated based on the shortest path principle; based on the relationship between the distance and the probability distribution, the probability corresponding to the first downstream road segment is determined according to the distance from the current road segment to the second node road segment through each downstream road segment.

[0115] The dimension of the shortest path principle can simulate the experience of the moving object in path planning in the geographical space (such as the road network).

[0116] Alternatively, the second probability can be denoted as: U(x) = e -γx where dist1 represents the distance from the current road segment rid i to the first downstream road segment rid j , and h(dist2) represents the distance from the downstream road segment rid jThe heuristic distance estimation to the destination D (the second node segment) is usually the Euclidean distance, i.e. the straight-line distance, in the absence of real-time data such as segment speed. U(x) is an equation that converts cost into utility. In this scheme, an equation similar to logistic regression is selected, i.e. the farther the distance, the lower the utility of selecting this downstream segment, and γ is an adjustable parameter. ∑U j The role of is to normalize the utility of all adjacent downstream segments of the current segment.

[0117] Of course, this is only an exemplary probability estimation method in the dimension of the shortest path principle, and the embodiment is not limited thereto.

[0118] Based on this, in the embodiment, the first probability and the second probability of the current segment entering the second node segment via the first downstream segment can be determined, and then the first probability and the second probability can be weighted and summed based on the sample penetration rate of the target area to determine the final probability of the current segment entering the second node segment via the first downstream segment.

[0119] When the sample penetration rate is 1 (i.e. the sample trajectory is full), the final probability of the current segment entering the second node segment via the first downstream segment is completely determined by the first probability generated by the sample trajectory dimension; and when there is no sample trajectory data, the final probability of the current segment entering the second node segment via the first downstream segment is completely determined by the second probability generated according to the experience of the shortest path principle.

[0120] Accordingly, in the embodiment, the probabilities of the current segment entering the second node segment via its adjacent downstream segments can be determined respectively, and based on these probabilities, the connection trajectory between the first node segment and the second node segment can be generated.

[0121] In actual application, starting from the first node segment, for the current segment, N downstream segments meeting the preset probability requirement can be selected from the adjacent downstream segments of the current segment as the next segment of the current segment until the second node segment is reached to generate the connection trajectory between the first node segment and the second node segment. Wherein, N is a positive integer.

[0122] Wherein, the requirement of the preset probability requirement can be set according to actual needs. For example, it can be set to select the N with the highest probability. For another example, the possible trajectory between the first node segment and the second node segment can also be represented as a probability string, i.e. a probability sequence formed according to the order and probability of each segment contained in the possible trajectory, and the preset probability requirement is limited from the perspective of the probability string, such as setting to select the N with the highest probability sum in the probability string, etc. The embodiment is not limited thereto.

[0123] In addition, in the embodiment, when the next road segment is selected for the current road segment, the number of trajectories corresponding to the first set of node road segments can also be considered, so as to generate a number of connection trajectories between the first node road segment and the second node road segment, which is consistent with the number of trajectories, as the goal to determine the number of next road segments selected for the current road segment. For example, if the number of trajectories corresponding to the first set of node road segments is 10, 10 connection trajectories can be generated between every two adjacent node road segments of the first set of node road segments, as the basis of the aforementioned sampling process.

[0124] It is worth noting that the above path planning scheme can also be applied in the extension of the trajectories corresponding to the at least one set of node road segments, so as to generate reachable trajectories between the trajectories corresponding to the at least one set of node road segments and the location hotspots.

[0125] In the above or the following embodiments, after the trajectories included in the target region in the target time period are generated, the moving object flow on the road segments other than the node road segments in the target region in the target time period can also be determined based on the trajectories included in the target region in the target time period.

[0126] In this way, the moving object flow on each road segment in the target region in the target time period can be obtained. This can serve as the data basis for subsequent processing links.

[0127] For example, in the traffic diversion processing link, the target time period to which the current time belongs can be determined; and a diversion scheme for the current time can be created according to the moving object flow on each road segment in the target region in the target time period and at least one time period after the target time period. For another example, in the traffic warning processing link, the target time period to which the current time belongs can be determined; and warning information for the current time can be created according to the moving object flow on each road segment in the target region in the target time period and at least one time period after the target time period.

[0128] The target region can also be controlled based on the trajectories included in the target region in the target time period.

[0129] For example, the queue length and other index parameters of the intersection can be calculated based on the trajectories included in the target region in the target time period. Since the trajectories included in the target region in the target time period generated in the embodiment are very comprehensive and accurate, the accuracy of these index parameters can be effectively improved. Thus, more accurate and more refined traffic control can be achieved.

[0130] In addition, from the perspective of deep analysis, the trajectories included in the target region in the target time period generated in the embodiment can continue to be analyzed by labeling, such as disassociating the POI information points, and finding out the travel activity information expressed by the common trajectories.

[0131] Further, from the perspective of traffic simulation application, the trajectories of the target region in the target period generated by the embodiment can be used as the input of micro-simulation. Micro-simulation refers to simulating the trajectories of each moving object, and the micro-simulation can be used to evaluate the influence of different traffic optimization schemes on the overall traffic condition.

[0132] In the above or the following embodiments, a trajectory generation interface can also be provided; the target region and the target period can be acquired in response to a parameter configuration operation occurring in the trajectory generation interface; and the trajectories of the target region in the target period generated can be output to the trajectory generation interface for display.

[0133] In actual application, the trajectory generation interface can be provided to a terminal device of a user and displayed in the terminal device of the user. The user can be an administrator of traffic control, an individual user, an enterprise user, or the like, and the embodiment is not limited in this regard.

[0134] In addition to performing a parameter configuration operation in the trajectory generation interface, the user can also perform a referencing operation on the trajectories of the target region in the target period generated by the embodiment based on various application schemes mentioned in the above embodiments, so as to apply the trajectories of the target region in the target period generated by the embodiment to a subsequent stage.

[0135] Figure 2 A schematic diagram of an application scenario provided by an embodiment of the present application.

[0136] As shown in Figure 2 , the embodiment of the present application can provide a trajectory generation interface, and acquire configuration parameters such as a target region and a target period based on the trajectory generation interface. According to the acquired configuration parameters, at least one group of node road segments of the target region in the target period can be summarized, Figure 2 which are represented by black dots. Each group of node road segments includes at least one node road segment, and also includes attribute information such as the order between the node road segments and the time distribution into each node road segment.

[0137] For example, a plurality of sample trajectories of the target region in the target period can be mapped to a road network structure to establish a mapping relationship between the sample trajectories and the node road segments, and then a plurality of samples can be clustered to summarize at least one type of sample trajectories of the target region in the target period, and the node road segments and the attribute information of the node road segments included in the at least one type of sample trajectories can be extracted respectively to construct at least one group of node road segments.

[0138] As shown in Figure 2 , it is assumed that the target region in the target period includes five groups of node road segments, Figure 2 two of which are shown in

[0139] For each group of node road segments, path planning can be performed to generate a corresponding trajectory for each group of node road segments.

[0140] Specifically, path planning can be performed between each two adjacent node road segments in each group of node road segments to determine the connection trajectory between each two node road segments of the cable, for example, Figure 2 In the first group of node road segments, two connection trajectories are generated between node road segment A and node road segment B, Figure 2 In the first group of node road segments, two connection trajectories are generated between node road segment A and node road segment B,

[0141] For each group of node road segments, sampling can be performed on the connection trajectory between each two node road segments of the cable to generate a corresponding trajectory for each group of node road segments.

[0142] The number of trajectories generated for each group of node road segments can be determined according to the traffic of the mobile objects collected by each node road segment of the target region in the target period, so that the number of trajectories generated for each group of node road segments is closer to the actual traffic observation value of the target region in the target period.

[0143] On this basis, the trajectories generated for each group of node road segments can be extended to extend these trajectories to the location hotspots of the target region in the target period. These location hotspots are determined based on the historical travel behavior of the target region in the target period, so the final trajectories generated by the extension process will be closer to the actual travel situation of the target region in the target period.

[0144] Accordingly, the full amount of trajectories of the target region in the target period can be generated, which conforms to the actual traffic observation value. Moreover, the generated trajectories meet the path planning experience and the rules of sample trajectories. Therefore, the travel data of the target region in the target period is more comprehensive and accurate.

[0145] It should be noted that in some of the processes described in the above embodiments and the accompanying drawings, a plurality of operations appear in a specific order, but it should be clearly understood that these operations can be executed or in parallel without the order in which they appear in this document. The serial numbers of the operations, such as 100, 101, 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 node road segments, downstream road segments, groups of node road segments, etc., and do not represent the order of precedence. Also, "first" and "second" are not different types.

[0146] Figure 3 A structural schematic diagram of a computing device is provided for another embodiment of the present application. As shown in Figure 3As shown, the computing device can include a memory 30 and a processor 31.

[0147] The memory 30 is used to store computer programs and can be configured to store other various data to support operations on the computing device. Examples of these data include instructions for any application or method operating on the computing device, contact data, phonebook data, messages, pictures, videos, etc.

[0148] The memory 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 memory, flash memory, magnetic disk or optical disk.

[0149] The processor 31 is coupled to the memory 30 and is used to execute computer programs in the memory for:

[0150] obtaining at least one set of node road segments contained by the target region in the target period;

[0151] respectively planning paths for the at least one set of node road segments to obtain trajectories respectively generated by the at least one set of node road segments;

[0152] generating a trajectory of the target region in the target period according to the trajectories respectively generated by the at least one set of node road segments.

[0153] In an optional embodiment, when the processor 31 generates the trajectory of the target region in the target period according to the trajectories respectively generated by the at least one set of node road segments, the processor 31 is configured to:

[0154] obtaining position hotspots respectively associated with the at least one set of node road segments, the position hotspots being position points outside the at least one set of node road segments in the target region;

[0155] extending the trajectories respectively generated by the at least one set of node road segments according to the position hotspots respectively associated with the at least one set of node road segments to obtain the trajectory of the target region in the target period.

[0156] In an optional embodiment, when the processor 31 extends the trajectories respectively generated by the at least one set of node road segments according to the position hotspots respectively associated with the at least one set of node road segments to obtain the trajectory of the target region in the target period, the processor 31 is configured to:

[0157] selecting a target position hotspot for the first trajectory from the position hotspots associated with the first set of node road segments for the first trajectory generated by the first set of node road segments;

[0158] extend the first trajectory based on the target position hotspot to obtain an accessible trajectory between the first trajectory and the target position hotspot;

[0159] connect the first trajectory and the accessible trajectory to obtain a trajectory of the target region in the target period;

[0160] The first group of node road segments is any one of the at least one group of node road segments, and the first trajectory is any one of the trajectories generated by the first group of node road segments.

[0161] In an optional embodiment, when the processor 31 respectively plans paths for the at least one group of node road segments to obtain the trajectories respectively generated by the at least one group of node road segments, the processor 31 is configured to:

[0162] obtain the number of trajectories respectively corresponding to the at least one group of node road segments;

[0163] respectively plan paths for the at least one group of node road segments to obtain the trajectories respectively generated by the at least one group of node road segments and meeting the number of trajectories thereof.

[0164] In an optional embodiment, when the processor 31 respectively plans paths for the at least one group of node road segments to obtain the trajectories respectively generated by the at least one group of node road segments and meeting the number of trajectories thereof, the processor 31 is configured to:

[0165] plan paths for the first group of node road segments to determine the connection trajectories between each two adjacent node road segments in the first group of node road segments;

[0166] sample the connection road segments between each two adjacent node road segments in the first group of node road segments according to the number of trajectories corresponding to the first group of node road segments as the sampling times, and each sampling result includes a target connection trajectory between each two adjacent node road segments in the first group of node road segments;

[0167] connect the target connection trajectory in each sampling result to the node road segments in the first group of node road segments to obtain the trajectories generated by the first group of node road segments and meeting the number of trajectories thereof;

[0168] The first group of node road segments is any one of the at least one group of node road segments.

[0169] In an optional embodiment, when the processor 31 plans paths for the first group of node road segments to determine the connection trajectories between each two adjacent node road segments in the first group of node road segments, the processor 31 is configured to:

[0170] For the adjacent first node road segment and second node road segment, the processor 31 calculates a probability that a current road segment enters the second node road segment through an adjacent downstream road segment of the current road segment, starting from the first node road segment;

[0171] determine a next road segment of the current road segment based on the probability until reaching a second node road segment, to generate a connected track between the first node road segment and the second node road segment;

[0172] wherein the first node road segment and the second node road segment are any two adjacent node road segments in the first group of node road segments.

[0173] In an optional embodiment, the processor 31, in calculating the probability of the current road segment entering the second node road segment through its adjacent downstream road segment, is configured to:

[0174] determine a first probability of the current road segment entering the second node road segment through its adjacent downstream road segment according to sample tracks in the target area;

[0175] determine a second probability of the current road segment entering the second node road segment through its adjacent downstream road segment based on the shortest path principle;

[0176] weight the first probability with a sample penetration rate corresponding to the target area as a weight, weight a difference obtained by subtracting the sample penetration rate from 1 as a weight of the second probability, and perform weighted summation on the first probability and the second probability to determine the probability of the current road segment entering the second node road segment through its adjacent downstream road segment.

[0177] In an optional embodiment, the processor 31, in determining the first probability of the current road segment entering the second node road segment through its adjacent downstream road segment according to sample tracks in the target area, is configured to:

[0178] for a first downstream road segment adjacent to the current road segment, based on a set of sample tracks in the target area that match the attribute information of the first group of node road segments, count a proportion of sample tracks entering the second node road segment from the current road segment through the first downstream road segment among sample tracks entering the second node road segment through the current road segment;

[0179] determine a first probability corresponding to the first downstream road segment according to the proportion;

[0180] wherein the first downstream road segment is any road segment among the downstream road segments adjacent to the current road segment.

[0181] In an optional embodiment, the processor 31, in determining the second probability of the current road segment entering the second node road segment through its adjacent downstream road segment based on the shortest path principle, is configured to:

[0182] for a first downstream road segment adjacent to the current road segment, calculate a distance from the current road segment to the second node road segment through the first downstream road segment based on the shortest path principle;

[0183] determine a second probability corresponding to the first downstream road segment according to distances from the current road segment to the second node road segment through each downstream road segment of the current road segment based on a relationship between the distance and the probability distribution;

[0184] wherein the first downstream road segment is any of the downstream road segments adjacent to the current road segment.

[0185] In an optional embodiment, the sample trajectory can be from navigation data.

[0186] In an optional embodiment, the processor 31, when determining the next road segment of the current road segment based on the probability until reaching the second node road segment for generating the connected trajectory between the first node road segment and the second node road segment, is configured to:

[0187] From the first node road segment, for the current road segment, selecting N downstream road segments meeting the preset probability requirement from the adjacent downstream road segments of the current road segment as the next road segment of the current road segment until reaching the second node road segment for generating the connected trajectory between the first node road segment and the second node road segment, N being a positive integer.

[0188] In an optional embodiment, the processor 31, when obtaining the at least one group of node road segments contained in the target region in the target period, is configured to:

[0189] Obtaining a plurality of sample trajectories in the target region and adaptive to the target period;

[0190] Clustering the plurality of sample trajectories based on the mapping relationship between the sample trajectories and the node road segments to obtain at least one type of sample trajectories;

[0191] Respectively extracting the node road segments and the attribute information of the node road segments contained in the at least one type of sample trajectories to generate a group of node road segments corresponding to each of the at least one type of sample trajectories.

[0192] In an optional embodiment, the processor 31, when obtaining the trajectory number corresponding to each of the at least one group of node road segments, is configured to:

[0193] Determining the number of sample trajectories contained in each of the at least one type of sample trajectories in a single period according to the number of sample trajectories and the number of periods contained in each of the at least one type of sample trajectories;

[0194] Based on the sample trajectory penetration rate corresponding to the target region, calculating the ratio of the number of sample trajectories contained in each of the at least one type of sample trajectories in a single period to the sample trajectory penetration rate as the trajectory number corresponding to each of the at least one group of node road segments.

[0195] In an optional embodiment, the processor 31, when obtaining the trajectory number corresponding to each of the at least one group of node road segments, is configured to:

[0196] Obtaining the mobile object flow collected by each node road segment in the target period;

[0197] The difference between the estimated flow and the mobile object flow collected by each node road section is minimized, and the number of trajectories corresponding to each node road section in at least one group is calculated under the condition of meeting the target.

[0198] The estimated flow is determined according to the number of trajectories corresponding to each node road section in at least one group and the node road sections contained in each node road section.

[0199] In an optional embodiment, when the processor 31 obtains the number of trajectories corresponding to each node road section in at least one group, it is used to:

[0200] According to the number of sample trajectories contained in at least one type of sample trajectory and the number of cycles, the number of sample trajectories contained in at least one type of sample trajectory under a single cycle is determined.

[0201] Based on the sample trajectory penetration rate corresponding to the target area, the ratio of the number of sample trajectories contained in at least one type of sample trajectory under a single cycle to the sample trajectory penetration rate is calculated as the reference trajectory number corresponding to each node road section in at least one group.

[0202] Obtain the mobile object flow collected by each node road section in the target period;

[0203] The difference between the estimated flow and the mobile object flow collected by each node road section is minimized, and the number of trajectories corresponding to each node road section in at least one group is calculated under the condition of meeting the target.

[0204] The estimated flow is determined according to the number of trajectories corresponding to each node road section in at least one group and the node road sections contained in each node road section.

[0205] In an optional embodiment, the node road section is a road section provided with a flow sensing device, and when the processor 31 obtains the mobile object flow collected by each node road section in the target period, it is used to:

[0206] Use the flow sensing device on each node road section to collect the mobile object flow of each node road section in the target period.

[0207] In an optional embodiment, the flow sensing device includes one or more of a camera or a ground inductive coil.

[0208] In an optional embodiment, the processor 31 is further used to:

[0209] Based on the trajectories contained in the target area in the target period, the mobile object flow on the road sections other than the node road sections in the target area in the target period is determined.

[0210] In an optional embodiment, the processor 31 is further used to:

[0211] determining a target period to which the current time belongs;

[0212] creating a traffic plan or warning information for the current time according to the traffic flow of the moving objects on each road segment in the target region in the target period and at least one period after the target period.

[0213] In an optional embodiment, the processor 31 is further configured to:

[0214] controlling the traffic of the target region based on the track contained in the target region in the target period.

[0215] In an optional embodiment, the processor 31 is further configured to:

[0216] providing a track generation interface;

[0217] obtaining the target region and the target period in response to a parameter configuration operation occurring in the track generation interface;

[0218] outputting the track contained in the target region in the target period to the track generation interface for display.

[0219] In an optional embodiment, the node road segment is a road segment provided with a flow sensing device.

[0220] Further, as shown in Figure 3 , the computing device further includes other components such as a communication component 32 and a power component 33. Figure 3 Only some components are shown in the figure schematically, and it does not mean that the computing device only includes the components shown in the figure. Figure 3

[0221] The communication component 32 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 WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component can be implemented based on near field communication (NFC) technology, radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra wide band (UWB) technology, Bluetooth (BT) technology or other technology to facilitate short-range communication.

[0222] The power component 33 provides power for various components of the device where the power component is located. The power component can include a power management system, one or more power sources, and other components associated with generating, managing and distributing power for the device where the power component is located.

[0223] ​It is worth noting that the technical details in the above related embodiments of the computing device can refer to the related descriptions in the foregoing trajectory generation method embodiments for brevity, and are not repeated here, but this should not cause loss of the protection scope of the present application.

[0224] Correspondingly, the embodiments of the present application also provide a computer readable storage medium storing a computer program, which can implement the steps performed by the computing device in the foregoing method embodiments when executed.

[0225] 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 take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0226] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products 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 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 apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the functions specified in the flowchart

[0227] These computer program instructions can also be stored in a computer readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the functions specified in the flowchart

[0228] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the functions specified in the flowchart

[0229] In one typical arrangement, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0230] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) having a common memory space employing many flash memory devices. The memory is an example of computer readable media.

[0231] Computer readable media includes permanent and non-permanent, removable and non-removable media 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 discs (DVDs) or other optical storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0232] 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 recited, but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0233] 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. The disclosure is intended to embrace all such modifications and changes that fall within the scope of the present application.

Claims

1. A trajectory generation method characterized by, include: Obtain at least one set of node road segments contained in the target area during the target time period; For any set of node road segments, the first node road segment and the second node road segment that are sequentially adjacent, starting from the first node road segment, the first probability of the current road segment entering the second node road segment via its adjacent downstream road segment is determined based on the sample trajectory in the target area. Based on the shortest path principle, determine the second probability that the current road segment will enter the second node road segment via its adjacent downstream road segment; The first probability and the second probability are fused to determine the probability that the current road segment enters the second node road segment via its adjacent downstream road segment; Based on the probability, determine the next road segment of the current road segment until the second node road segment is reached, so as to generate the connection trajectory between the first node road segment and the second node road segment; By connecting the first node segment and the second node segment through the connecting trajectory, path planning is performed on each of the node segments to obtain the trajectory generated by the node segments. Based on the trajectories generated by each of the at least one set of node road segments, the trajectory contained in the target area during the target time period is generated.

2. The method of claim 1, wherein, Based on the trajectories generated by each of the at least one set of node road segments, the trajectory of the target area during the target time period is generated, including: Obtain at least one set of location hotspots associated with each set of node road segments, wherein the location hotspots are location points outside the at least one set of node road segments within the target area; Based on the location hotspots associated with each of the at least one set of node road segments, the trajectories generated by each of the at least one set of node road segments are extended to obtain the trajectory contained in the target area during the target time period.

3. The method of claim 2, wherein, Based on the location hotspots associated with each of the at least one set of node road segments, the trajectories generated by each of the at least one set of node road segments are extended to obtain the trajectory contained in the target area during the target time period, including: For the first trajectory generated by the first group of node segments, a target location hotspot is selected for the first trajectory from the location hotspots associated with the first group of node segments; Based on the target location hotspot, the first trajectory is extended to obtain a reachable trajectory between the first trajectory and the target location hotspot; Connect the first trajectory and the reachable trajectory to obtain a trajectory of the target area during the target time period; Wherein, the first group of node segments is any one of the at least one group of node segments, and the first trajectory is any one of the trajectories generated by the first group of node segments.

4. The method of claim 1, wherein, Also includes: Obtain the number of trajectories corresponding to each of the at least one set of node segments; Path planning is performed on each of the at least one set of node road segments to obtain trajectories that meet the required number of trajectories for each of the at least one set of node road segments.

5. The method of claim 4, wherein, Path planning is performed on each of the at least one set of node road segments to obtain trajectories that meet the required number of trajectories for each of the at least one set of node road segments, including: The number of trajectories corresponding to the first set of node road segments is taken as the sampling times, and the connecting road segments between each two adjacent node road segments in the first set of node road segments are sampled, and each sampling result includes a target connecting trajectory between each two adjacent node road segments in the first set of node road segments; The target connecting trajectory in each sampling result is connected with the node road segment in the first set of node road segments respectively, and the trajectories conforming to the number of trajectories of the first set of node road segments are obtained. The first set of node road segments is any one of the at least one set of node road segments.

6. The method of claim 1, wherein, The first probability and the second probability are fused to determine the probability of the current road segment entering the second node road segment through the adjacent downstream road segment, including: The sample trajectory penetration rate corresponding to the target region is taken as the weight of the first probability, and the difference obtained by subtracting the sample trajectory penetration rate from 1 is taken as the weight of the second probability, the first probability and the second probability are weighted and summed to determine the probability of the current road segment entering the second node road segment through the adjacent downstream road segment. The sample trajectory penetration rate refers to the proportion of the number of sample trajectories in the target region to the total amount of trajectories.

7. The method of claim 1, wherein, The first probability of the current road segment entering the second node road segment through the adjacent downstream road segment is determined according to the sample trajectories in the target region, including: For the first downstream road segment adjacent to the current road segment, based on the set of sample trajectories in the target region that match the attribute information of the first set of node road segments, the proportion of sample trajectories entering the second node road segment from the current road segment through the first downstream road segment to sample trajectories entering the second node road segment through the current road segment is counted. According to the proportion, the first probability corresponding to the first downstream road segment is determined. The first downstream road segment is any road segment adjacent to the current road segment; the first set of node road segments is any one of the at least one set of node road segments.

8. The method of claim 1, wherein, The second probability of the current road segment entering the second node road segment through the adjacent downstream road segment is determined based on the shortest path principle, including: For the first downstream road segment adjacent to the current road segment, the distance from the current road segment to the second node road segment through the first downstream road segment is calculated based on the shortest path principle. Based on the relationship between distance and probability distribution, the second probability corresponding to the first downstream road segment is determined according to the distance from the current road segment to the second node road segment through each downstream road segment. The sample trajectories come from navigation data.

9. The method of claim 6, wherein, Based on the probability, the next road segment of the current road segment is determined until the second node road segment is reached to generate the connecting trajectory between the first node road segment and the second node road segment, including:

10. The method of claim 1, wherein, Starting from the first node road segment, for the current road segment, from the adjacent downstream road segments, N downstream road segments meeting the preset probability requirement are selected as the next road segment of the current road segment until the second node road segment is reached to generate the connecting trajectory between the first node road segment and the second node road segment, N is a positive integer. ​ 11. The method of claim 4, wherein, The at least one group of node road segments contained in the target region in the target period comprises: Obtaining a plurality of sample trajectories in the target region and adaptive to the target period; Based on the mapping relationship between the sample trajectories and the node road segments, the plurality of sample trajectories are clustered to obtain at least one type of sample trajectories; Respectively extracting the node road segments and attribute information of the node road segments contained in the at least one type of sample trajectories to generate a group of node road segments corresponding to each of the at least one type of sample trajectories.

12. The method of claim 11, wherein, The obtaining of the trajectory quantity corresponding to each of the at least one group of node road segments comprises: According to the sample trajectory quantity and the period quantity contained in each of the at least one type of sample trajectories, the sample trajectory quantity contained in each of the at least one type of sample trajectories in a single period is determined; Based on the sample trajectory penetration rate corresponding to the target region, the ratio of the sample trajectory quantity contained in each of the at least one type of sample trajectories in a single period to the sample trajectory penetration rate is calculated as the trajectory quantity corresponding to each of the at least one group of node road segments; The sample trajectory penetration rate refers to the proportion of the number of sample trajectories in the target region to the total amount of trajectories.

13. The method of claim 11, wherein, The obtaining of the trajectory quantity corresponding to each of the at least one group of node road segments comprises: Obtaining the mobile object flow collected by each node road segment in the target period; Taking the minimum difference between the estimated flow of each node road segment and the mobile object flow collected by the node road segment as the target, the trajectory quantity corresponding to each of the at least one group of node road segments under the condition of satisfying the target is calculated; The estimated flow is determined according to the trajectory quantity corresponding to each of the at least one group of node road segments and the node road segments contained in each group.

14. The method of claim 11, wherein, The obtaining of the trajectory quantity corresponding to each of the at least one group of node road segments comprises: According to the sample trajectory quantity and the period quantity contained in each of the at least one type of sample trajectories, the sample trajectory quantity contained in each of the at least one type of sample trajectories in a single period is determined; Based on the sample trajectory penetration rate corresponding to the target region, the ratio of the sample trajectory quantity contained in each of the at least one type of sample trajectories in a single period to the sample trajectory penetration rate is calculated as the reference trajectory quantity corresponding to each of the at least one group of node road segments; Obtaining the mobile object flow collected by each node road segment in the target period; Taking the minimum sum value between the difference between the estimated flow of each node road segment and the mobile object flow collected by the node road segment and the difference between the trajectory quantity corresponding to each of the at least one group of node road segments and the reference trajectory quantity as the target, the trajectory quantity corresponding to each of the at least one group of node road segments is calculated; The estimated flow is determined according to the trajectory quantity corresponding to each of the at least one group of node road segments and the node road segments contained in each group; the sample trajectory penetration rate refers to the proportion of the number of sample trajectories in the target region to the total amount of trajectories.

15. The method according to claim 13 or 14, characterized in that, The node road segment is a road segment provided with a flow sensing device, and the obtaining of the mobile object flow collected by each node road segment in the target period comprises: Using the flow sensing device on each node road segment, the mobile object flow of each node road segment in the target period is collected respectively.

16. The method of claim 15, wherein, The flow-aware device comprises one or more of a camera or a ground inductive coil.

17. The method of claim 1, wherein, Further comprising: determining, based on the trajectory of the target area in the target period, a flow of moving objects on road segments other than node road segments within the target area in the target period.

18. The method of claim 17, wherein, Further comprising: determining a target period to which a current time belongs; creating a traffic diversion scheme or warning information for the current time according to the flow of moving objects on road segments within the target area in the target period and at least one period after the target period.

19. The method of claim 1, wherein, Further comprising: controlling traffic of the target area based on the trajectory of the target area in the target period.

20. The method of claim 1, wherein, Further comprising: providing a trajectory generation interface; in response to a parameter configuration operation occurring in the trajectory generation interface, obtaining the target area and the target period; outputting the generated trajectory of the target area in the target period to the trajectory generation interface for display.

21. A computing device, comprising: comprising a memory and a processor; the memory is configured to store one or more computer instructions; the processor is coupled to the memory and configured to execute the one or more computer instructions to execute the trajectory generation method of any one of claims 1-20.

22. A computer-readable storage medium storing computer instructions, wherein, when the computer instructions are executed by one or more processors, the one or more processors are caused to execute the trajectory generation method of any one of claims 1-20.

Images