Vehicle detour identification detection method and system based on trajectory similarity

Abstract

The invention discloses a vehicle detour identification detection method and system based on trajectory similarity, relates to the technical field of vehicles, and solves the technical problem of low accuracy of detour detection through single-point offset in a vehicle detour detection method in the prior art. Comprises: determining path recommendation update points based on a traffic network map; acquiring an initial position and a target position; generating a plurality of recommended path groups based on the initial position and the target position; obtaining driving data of the vehicle in real time, and generating a plurality of stages of recommended path groups based on the driving data and the path recommendation update points; a driving track is obtained, driving is segmented based on the path recommendation updating points, and a plurality of segmented tracks are obtained; detour risk detection is carried out based on the plurality of segmented tracks and the recommended path groups of all the stages; the detour risk detection comprises track risk detection and travel risk detection; analyzing the matching degree of the segmented track and the recommended path group; multi-section detour risk analysis is realized, and the detour analysis accuracy is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of vehicles, and in particular to a vehicle detour identification and detection method and system based on trajectory similarity. BACKGROUND

[0002] Operational vehicles, such as taxis or online car-hailing vehicles, have always been one of the important choices for urban travel. However, in order to obtain more profits, some drivers of operational vehicles will choose longer routes to travel in road sections that are not familiar to passengers. This detour behavior not only causes passengers to spend unnecessary money, but also seriously affects the passenger's ride experience, thereby reducing the number of people choosing to travel by operational vehicles. This not only brings inconvenience to passengers, but also causes economic losses to operational companies; in the modern vehicle dispatching and supervision scene, supervisors and dispatching systems need to monitor the vehicle trajectory in real time to ensure that the vehicle travels according to the established route, completes the task on time, and avoids detours or illegal operations.

[0003] However, the existing vehicle monitoring method usually relies on electronic fence judgment or single-point deviation detection, and these methods have obvious limitations: first, single-point deviation detection cannot accurately distinguish between temporary deviation of the vehicle and real detour behavior; analysis methods based only on distance or single-point deviation lack the ability to compare the overall trajectory, and easily ignore the overall driving mode of the trajectory; secondly, when facing long-distance trajectories or complex road environments, traditional methods often appear misjudgment or omission. In addition, most existing solutions lack automatic and intelligent processing capabilities, requiring frequent manual intervention or combining complex rule configurations, resulting in low dispatching efficiency and increased management costs; therefore, a vehicle detour identification and detection method and system based on trajectory similarity are needed. SUMMARY

[0004] The application provides a vehicle detour identification and detection method and system based on trajectory similarity, which solves the technical problem of low accuracy of existing vehicle detour detection methods through single-point deviation.

[0005] To achieve the above purpose, the application adopts the following technical solutions: In a first aspect, a vehicle detour identification and detection method based on trajectory similarity is provided, comprising: Obtaining a traffic road network map, determining a path recommendation update point based on the traffic road network map; Obtaining a starting position and a target position; generating a plurality of recommended path groups based on the starting position and the target position; Obtaining real-time driving data of the vehicle, generating a plurality of stage recommended path groups based on the driving data and the path recommendation update point; Obtain a driving track, segment the driving based on the path recommendation update point to obtain a plurality of segmented tracks; perform a detour risk detection based on the plurality of segmented tracks and the plurality of stage recommendation path groups; the detour risk detection includes track risk detection and travel risk detection.

[0006] Based on the above technical solutions, in the vehicle detour identification and detection method and system based on track similarity provided in the application, the traffic road network map is obtained, the path recommendation update point is determined according to the traffic road network map; the starting position and the target position are obtained; a plurality of recommendation path groups are generated according to the starting position and the target position; the driving data of the vehicle is obtained in real time, and a plurality of stage recommendation path groups are generated according to the driving data and the path recommendation update point; the driving track is obtained, the driving is segmented based on the path recommendation update point to obtain a plurality of segmented tracks; the detour risk detection is performed based on the plurality of segmented tracks and the plurality of stage recommendation path groups; the detour risk detection includes track risk detection and travel risk detection; the matching degree of the segmented track and the recommendation path group is analyzed; the multi-section detour risk analysis is realized, and the accuracy of the detour analysis is increased.

[0007] In combination with the above first aspect, in a possible implementation manner, the path recommendation update point is determined based on the traffic road network map, including: The node positions corresponding to the intersection nodes in the traffic road network map are obtained, and the node positions are recorded as the path recommendation update points; the traffic road network map is a road network map composed of various roads, and the intersection nodes are intersection points of different roads.

[0008] In combination with the above first aspect, in a possible implementation manner, the plurality of recommendation path groups are generated based on the starting position and the target position, including: A plurality of feasible paths are generated based on the starting position and the target position, the feasible path being a passable path from the current position of the automobile to the target position within a set range; it can be understood that when the current position of the automobile is the starting position, the feasible path is a passable path from the starting position to the target position; the set range is generally a circular region with the straight line center of the starting position and the target position as the center and the straight line distance between the starting position and the target position as the radius; a plurality of intersection nodes in each feasible path are obtained, and the node driving distance and the node driving time between adjacent intersection nodes are calculated; the driving time is The sum of the node driving distances between each adjacent intersection node of the feasible path is calculated to obtain the driving distance of the feasible path; the sum of the node driving times between each adjacent intersection node of the feasible path is calculated to obtain the driving time of the feasible path; The distance threshold and the time threshold are obtained, and the feasible path with a driving distance less than the distance threshold and a driving time less than the time threshold is recorded as a recommended path; the distance threshold is a set multiple of the average driving distance from the starting position to the target position in the historical data, for example, the distance threshold is 1.1 times of the average driving distance; the time threshold is a set multiple of the average driving time from the starting position to the target position in the historical data; for example, the time threshold is 1.15 times of the average driving time. The overlapping ratio between each recommended path is calculated, and when the overlapping ratio is greater than a set overlapping ratio threshold, the two recommended paths corresponding to the overlapping ratio are divided into the same recommended path group; it can be understood that the overlapping ratio is the ratio of the overlapping part of the two different paths to the length of the shortest path in the two paths; the overlapping ratio threshold is generally set to 80%.

[0009] In combination with the first aspect, in a possible implementation manner, one of the obtaining manners of the node driving time includes: The historical data of each time period between each intersection node is obtained; the average passing time of each vehicle in the historical data is extracted, and the average passing time is recorded as the passing time of the corresponding time period between the intersection nodes; The time period of arriving at the intersection node and the passing time between the intersection node and the adjacent next intersection node in the time period are obtained, and the passing time is recorded as the node driving time between the two intersection nodes; the time period of arriving at the intersection node is the starting time superimposed on the time period corresponding to each node driving time before the intersection node.

[0010] In combination with the first aspect, in a possible implementation manner, the generating of the several-stage recommended path group based on the driving data and the path recommendation update point includes: The target driving path is obtained, and the target driving path is a recommended path currently being driven; that is, a recommended path selected by the driver or passenger from each recommended path group; The positioning information and the forward direction in the driving data are obtained; the next recommended update point on the target driving path is determined based on the positioning information and the forward direction; The several feasible paths between the node position of the recommended update point and the target position are obtained, and the driving distance and the driving time of each feasible path are calculated; the initial distance threshold and the initial time threshold of the recommended update point are obtained; the initial distance threshold is the product of the distance threshold adjustment multiple and the initial average driving distance, and the initial time threshold is the product of the time threshold adjustment multiple and the initial average driving time; the node detour risk of the recommended update point is generated based on the initial distance threshold and the initial time threshold corresponding to the several feasible paths, and the driving distance and the driving time. The distance threshold adjustment multiple and the time threshold adjustment multiple are corrected based on the node detour risk, and a corrected node distance threshold and a corrected node time threshold are generated; When the driving distance of the feasible path is less than the node distance threshold and the driving time is less than the node time threshold, the feasible path is recorded as a stage recommended path of the recommended node; The overlap ratio between each stage recommended path is calculated, and when the overlap ratio is greater than a set overlap ratio threshold, the two recommended paths corresponding to the overlap ratio are divided into the same stage recommended path group.

[0011] In combination with the first aspect, in a possible implementation manner, the node detour risk of the recommended update point is generated based on the initial distance threshold and the initial time threshold corresponding to the feasible path, and the driving distance and the driving time, and includes: When the driving distance of the feasible path is greater than the set initial distance threshold and the driving time is greater than the set initial time threshold, the feasible path is recorded as a detour path; When the driving distance of the feasible path is less than or equal to the set initial distance threshold, or the driving time is less than or equal to the set initial time threshold, the feasible path is marked as a normal path; The node detour risk of the corresponding node is obtained by substituting the initial distance threshold and the initial time threshold into a set detour risk quantification function based on the driving distance and the driving time of the detour path and the normal path; one expression form of the detour risk quantification function includes: ; Wherein, is the node detour risk; is a driving distance risk quantification function, is an array composed of the driving distances of all the detour paths, i is the number of the detour path, i=1, 2, …, I, I is the total number of the driving distances in the array, that is, the total number of the detour paths; is an array composed of the driving distances of all the normal paths, j is the number of the normal path, j is the number of the detour path, j=1, 2, …, J, J is the total number of the driving distances in the array, that is, the total number of the normal paths; is the initial distance threshold; is a driving time risk quantification function, is an array composed of the driving times of all the detour paths, is an array composed of the driving times of all the normal paths, is the initial time threshold.

[0012] In conjunction with the first aspect above, in one possible implementation, the step of correcting the distance threshold adjustment factor and the time threshold adjustment factor based on node detour risk, and generating corrected node distance thresholds and node time thresholds, includes: The node detour risk is denoted as The distance threshold adjustment factor is denoted as The time threshold adjustment factor is denoted as ; Through formula The corrected distance threshold adjustment factor was calculated. ; Through formula The corrected time threshold adjustment factor was calculated. ; The product of the distance threshold adjustment factor and the initial average driving distance is recorded as the corrected node distance threshold. The product of the time threshold adjustment factor and the initial average travel time is denoted as the corrected node time threshold.

[0013] In conjunction with the first aspect above, in one possible implementation, the segmentation of trajectory data based on path recommendation update points to obtain several segmented trajectory data includes: Extract the driving trajectory from the trajectory data; the driving trajectory is the trajectory already traveled in this trip; obtain several path recommendation update points in the traffic network map, and segment the driving trajectory based on the path recommendation update points to obtain several segmented trajectories, the segmented trajectories being the driving trajectories between adjacent recommendation update points.

[0014] In conjunction with the first aspect above, in one possible implementation, the detour risk detection includes trajectory risk detection and travel risk detection; The trajectory risk detection includes: obtaining several stage recommended path groups corresponding to the path recommendation update point at the start of the segmented trajectory; extracting the number of stage recommended paths in the several stage recommended path groups that do not contain the segmented trajectory, and the sum of stage recommended paths in each stage recommended path group; recording the ratio of the number to the sum as the segmented trajectory risk of the segmented trajectory; sequentially calculating the segmented trajectory risk of each segmented trajectory, and calculating the average value of each segmented trajectory risk; recording the average value as the trajectory risk of the driving trajectory; and generating a detour alarm when the trajectory risk is greater than a set trajectory risk threshold. The trip risk detection comprises: acquiring the track length of each segmented track, summing up the track length of each track to obtain a traveled distance, and acquiring a recommended update node at the terminal of the last segmented track; acquiring a plurality of stage recommended path groups corresponding to the recommended update node, selecting the maximum length in each stage recommended path in each recommended path group, and recording the maximum length as an untraveled distance; when the sum of the traveled distance and the untraveled distance is greater than a distance threshold of a set multiple; the set multiple is a number greater than 1 set by an expert through experience, and the set multiple in this embodiment is 1.2 times; and generating a detour warning. The detour risk prediction further comprises: acquiring a node detour risk corresponding to a starting path recommended update point of the segmented track; calculating an average value of each node detour risk; when the average value is greater than a set node risk threshold, the node risk threshold is set by expert experience; and generating a detour warning, which indicates that the driver is always driving on a road segment with a high detour risk, and the possibility of detour is greater.

[0015] In a second aspect, the present application provides a vehicle detour identification and detection device based on track similarity, comprising: a processor and a storage medium; the storage medium comprises instructions, and the processor is configured to run the instructions to implement the method described in the first aspect and any possible implementation manner of the first aspect. The vehicle detour identification and detection device based on track similarity can be an electronic device or a chip in an electronic device.

[0016] In a third aspect, the present application provides a vehicle detour identification and detection system based on track similarity, comprising: a data acquisition module, a data analysis module and a risk warning module; wherein, The data acquisition module comprises a network acquisition unit and a vehicle terminal acquisition unit. The network acquisition unit is configured to acquire a traffic road network map, a starting position and a target position. The vehicle terminal acquisition unit is configured to acquire driving data and a driving track of a vehicle during driving. The data analysis module comprises a path recommendation unit and a risk assessment unit. The path recommendation unit is configured to determine a path recommended update point based on the traffic road network map, generate a plurality of recommended path groups based on the starting position and the target position, and generate a plurality of stage recommended path groups based on the driving data and the path recommended update point. The risk assessment unit is configured to segment driving based on the path recommended update point to obtain a plurality of segmented tracks, and perform detour risk detection based on the plurality of segmented tracks and the plurality of stage recommended path groups; the detour risk detection comprises track risk detection and trip risk detection. The risk warning module is configured to acquire a detour warning and a detour warning, and perform warning.

[0017] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a trajectory similarity-based vehicle detour identification and detection device, cause the trajectory similarity-based vehicle detour identification and detection device to perform the method described in the first aspect and any possible implementation thereof.

[0018] Fifthly, this application provides a computer program product containing instructions that, when the computer program product is run on a vehicle detour identification and detection device based on trajectory similarity, causes the vehicle detour identification and detection device based on trajectory similarity to perform the method described in the first aspect and any possible implementation thereof.

[0019] This application provides a vehicle detour identification and detection method based on trajectory similarity. It can acquire a traffic network map, determine route recommendation update points based on the map, obtain the starting and target positions, generate several recommended path groups based on the starting and target positions, acquire vehicle driving data in real time, generate several stage recommended path groups based on the driving data and route recommendation update points, acquire the driving trajectory, segment the driving based on the route recommendation update points to obtain several segmented trajectories, and perform detour risk detection based on the segmented trajectories and the recommended path groups at each stage. The detour risk detection includes trajectory risk detection and trip risk detection. Through matching degree analysis between segmented trajectories and recommended path groups, multi-segment detour risk analysis is achieved, increasing the accuracy of detour analysis.

[0020] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating the steps of the vehicle detour identification and detection method in this application; Figure 2 This is a schematic diagram of the module connections for the vehicle detour identification and detection system in this application. Detailed Implementation

[0023] The technical solutions of this application will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] Please see Figure 1 The first aspect of this application provides a vehicle detour identification and detection method based on trajectory similarity, comprising: Obtain a traffic network map and determine the route recommendation update point based on the traffic network map; Obtain the starting and target positions; generate several recommended path groups based on the starting and target positions; Real-time acquisition of vehicle driving data, and generation of several stages of recommended path groups based on driving data and path recommendation update points; The driving trajectory is obtained, and the driving is segmented based on the path recommendation update points to obtain several segmented trajectories; detour risk detection is performed based on several segmented trajectories and recommended path groups at each stage; the detour risk detection includes trajectory risk detection and trip risk detection.

[0025] Based on the above technical solutions, the vehicle detour identification and detection method and system based on trajectory similarity provided in this application involves: acquiring a traffic network map and determining path recommendation update points based on the traffic network map; acquiring the starting position and target position; generating several recommended path groups based on the starting position and target position; acquiring vehicle driving data in real time and generating several stage recommended path groups based on the driving data and path recommendation update points; acquiring the driving trajectory and segmenting the driving based on the path recommendation update points to obtain several segmented trajectories; performing detour risk detection based on several segmented trajectories and each stage recommended path group; the detour risk detection includes trajectory risk detection and trip risk detection; and achieving multi-segment detour risk analysis through matching degree analysis between segmented trajectories and recommended path groups, thereby increasing the accuracy of detour analysis. Simultaneously, through staged path recommendation and dynamic path filtering, detours caused by congestion on pre-selected roads are avoided, i.e., reasonable detour analysis is not mistaken for detour behavior, further increasing the accuracy of detour identification.

[0026] In one possible implementation, determining the path recommendation update point based on the traffic network map includes: extracting each intersection node in the traffic network map; obtaining the node position corresponding to each intersection node, and recording the node position as the path recommendation update point, wherein the traffic network map is a road network map composed of various roads, and the intersection node is the intersection of different roads.

[0027] In one possible implementation, several recommended path groups are generated based on the starting position and the target position, including: generating several feasible paths based on the starting position and the target position. Feasible paths are traversable paths from the current vehicle position to the target position within a defined range. It can be understood that when the vehicle's current position is the starting position, a feasible path is a traversable path from the starting position to the target position. The defined range is generally a circular area centered on the straight-line center between the starting position and the target position, with a radius equal to the straight-line distance between the starting position and the target position. Several intersection nodes in each feasible path are obtained, and the node travel distance and node travel time between adjacent intersection nodes are calculated. The travel time is...

[0028] The travel distance of the feasible path is obtained by summing the travel distances between each adjacent intersection node of the feasible path; the travel time of the feasible path is obtained by summing the travel times between each adjacent intersection node of the feasible path. Obtain distance and time thresholds, and record feasible paths whose travel distance is less than the distance threshold and whose travel time is less than the time threshold as recommended paths; the distance threshold is a set multiple of the average travel distance from the starting position to the target position in historical data, such as 1.1 times the average travel distance; the time threshold is a set multiple of the average travel time from the starting position to the target position in historical data, such as 1.15 times the average travel time, etc. Calculate the overlap ratio between each recommended path. When the overlap ratio is greater than a set overlap ratio threshold, the two recommended paths corresponding to the overlap ratio are grouped into the same recommended path group. It can be understood that the overlap ratio is the proportion of the overlapping part of two different paths to the length of the shortest path between the two paths. The overlap ratio threshold is generally set to 80%.

[0029] It is understood that in this embodiment, when displaying routes, the recommended route with the shortest travel time from each recommended route group will be selected as the representative route of that recommended route group for display. That is, only one route is displayed for each recommended route group, and the other recommended routes in the recommended route group are just alternative routes to the alternative route. This way, the number of routes displayed will not be large, the display will be simple, and it will be convenient for drivers and users to select. At the same time, in the event of traffic jams or other situations, an alternative recommended route can be quickly selected from the recommended route group corresponding to the selected recommended route, thereby improving the user experience.

[0030] In one possible implementation, one way to obtain node travel time includes: obtaining historical data for each time period between each intersection node; extracting the average travel time of each vehicle from the historical data, and recording the average travel time as the travel time between the intersection nodes for the corresponding time period; Obtain the time period for arriving at the intersection node, and the travel time between the intersection node and the next adjacent intersection node within the time period, and record the travel time as the node travel time between the two intersection nodes; the time period for arriving at the intersection node is the time period corresponding to the start time plus the travel time of each node before the intersection node; the node travel time between the first intersection node and the starting position is the travel time from the starting position to the first intersection node; it can be obtained by multiplying the ratio of the distance from the starting position to the first intersection node to the distance between the previous intersection node and the first intersection node on the road where the starting position is located by the travel time between the previous intersection node and the first intersection node in that time period.

[0031] Specifically, if a day is divided into 24 time periods, each lasting one hour, and the journey begins at 9:45 from the first intersection, the time period between the first and second intersections is from 9:00 to 10:00. If the travel time between the first and second intersections exceeds 15 minutes, then the time period between the second and third intersections is from 10:00 to 11:00.

[0032] In one possible implementation, several stages of recommended route groups are generated based on driving data and route recommendation update points, including: obtaining a target driving route, which is the currently driving recommended route; that is, a recommended route selected by the driver or passenger from each recommended route group; Retrieve location information and direction of travel from the driving data; determine the next recommended update point on the target driving path based on the location information and direction of travel; Several feasible paths are obtained between the node location of the recommended update point and the target location, and the travel distance and travel time of each feasible path are calculated. An initial distance threshold and an initial time threshold are obtained for the recommended update point. The initial distance threshold is the product of a distance threshold adjustment factor and an initial average travel distance. The initial time threshold is the product of a time threshold adjustment factor and an initial average travel time. The initial average travel distance is the average travel distance from the node location corresponding to the recommended update point in historical data to the target location. For example, if the initial distance threshold is 1.1 times the initial average travel distance, the distance threshold adjustment factor is 1.1. The initial average travel time is the average travel time from the node location corresponding to the recommended update point in historical data to the target location. For example, if the initial time threshold is 1.15 times the initial average travel time, the time threshold adjustment factor is 1.15. Based on the initial distance threshold and initial time threshold corresponding to several feasible paths, as well as the travel distance and travel time, the node detour risk of the recommended update point is generated. The distance threshold adjustment factor and time threshold adjustment factor are corrected based on the node detour risk, and the corrected node distance threshold and node time threshold are generated. When the travel distance of a feasible path is less than the node distance threshold and the travel time is less than the node time threshold, the feasible path is recorded as the stage recommended path of the recommended node. Calculate the overlap ratio between recommended paths at each stage. When the overlap ratio is greater than a set overlap ratio threshold, the two recommended paths corresponding to the overlap ratio are grouped into the same stage recommended path group.

[0033] The stage route recommendation group is displayed in the same way as the above-mentioned recommendation route group, and can realize real-time updates of the recommendation route group, further ensuring that alternative routes can be quickly selected when traffic jams or other traffic situations occur, thus improving the user experience.

[0034] In one possible implementation, the node detour risk of the recommended update point is generated based on the initial distance threshold and initial time threshold corresponding to several feasible paths, as well as the travel distance and travel time. This includes: when the travel distance corresponding to a feasible path is greater than the set initial distance threshold and the travel time is greater than the set travel time threshold, the feasible path is recorded as a detour path. When the travel distance corresponding to a feasible path is less than or equal to the set initial distance threshold, or the travel time is less than or equal to the set initial time threshold, the feasible path is marked as a normal path. Based on the travel distance and time of the detour route and the normal route, and substituting initial distance thresholds and initial time thresholds into a predefined detour risk quantification function, the detour risk of the corresponding node is obtained; one expression of the detour risk quantification function includes: ; in, Risk of node detours; For the risk quantification function of driving distance, The array consists of the driving distances corresponding to each detour route, where i is the number of the detour route, i=1,2,...,I, and I is the total number of driving distances in the array, which is also the total number of detour routes. The array consists of the driving distances corresponding to each normal path, where j is the number of the normal path and j is the number of the detour path, j=1,2,...,J, and J is the total number of driving distances in the array, which is also the total number of normal paths. This is the initial distance threshold; For the risk quantification function of travel time, An array consisting of the travel times for each detour route. An array consisting of the travel times for each normal route. This is the initial time threshold; One expression of the driving distance risk quantification function is as follows: ; in, The i-th travel distance in the array consisting of the travel distances corresponding to each detour route; This is the j-th driving distance in the array consisting of driving distances corresponding to each normal path; it is understood that the driving distance risk quantification function provided above is only one form of expression; The greater the difference between the average driving distance of all detours and the average driving distance of the normal route for a recommended update point, the longer the detours or the greater the number of detours compared to the normal route. This indicates that the distance after the detour from the recommended update point is significantly increased, or that there are more detour options available. In this case, the detour risk of the recommended update node is set to be relatively high. One expression of the driving time risk quantification function is as follows: ; in, The i-th travel time in the array consisting of the travel times for each detour route; The j-th travel time is the number in the array of travel times for each normal path. It should be understood that the travel distance risk quantification function provided above is only one form of expression. Understandably, the logical relationship of the above formula is basically the same as that of the driving distance risk quantification function, except that the risk of detours is analyzed synchronously at the time level. That is, the greater the difference between the average driving time of all detours and the average driving time of the normal route for a certain route recommendation update point, the longer the detours take, or the more detours there are than the number of normal routes. This indicates that the travel time after the detour from the recommendation update point is longer, or that there are more detour options available. In this case, the detour risk of the recommendation update node is set higher.

[0035] In one possible implementation, the distance threshold adjustment factor and time threshold adjustment factor are corrected based on the node detour risk, and the corrected node distance threshold and node time threshold are generated, including: denoting the node detour risk as... The distance threshold adjustment factor is denoted as The time threshold adjustment factor is denoted as ; Through formula The corrected distance threshold adjustment factor was calculated. ; Through formula The corrected time threshold adjustment factor was calculated. ; The product of the distance threshold adjustment factor and the initial average driving distance is recorded as the corrected node distance threshold. The product of the time threshold adjustment factor and the initial average travel time is denoted as the corrected node time threshold.

[0036] The greater the risk of detours, the better the conditions for drivers to detour at that node, and the greater the possibility of detours. In this case, it is necessary to narrow down the selection range of recommended routes to avoid providing too many recommended routes to drivers to facilitate detours, while also improving the accuracy of subsequent detour analysis and evaluation.

[0037] In one possible implementation, the trajectory data is segmented based on path recommendation update points to obtain several segmented trajectory data; this includes: extracting the driving trajectory from the trajectory data; the driving trajectory is the trajectory already traveled in this trip; obtaining several path recommendation update points in the traffic network map, and segmenting the driving trajectory based on the path recommendation update points to obtain several segmented trajectories, the segmented trajectories being the driving trajectories between adjacent recommendation update points; specifically, there are existing path recommendation update points A, B, C, and D; the driving trajectory passes through path recommendation update points A, B, C, and D in sequence. At this time, the driving trajectory portion between path recommendation update points A and B, the driving trajectory portion between path recommendation update points B and C, and the driving trajectory portion between path recommendation update points C and D are all segmented trajectories.

[0038] In one possible implementation, detour risk detection includes trajectory risk detection and trip risk detection. Trajectory risk detection includes: obtaining several recommended path groups corresponding to the path recommendation update point at the start of the segmented trajectory; extracting the number of recommended paths in the recommended path groups that do not contain the segmented trajectory, and the sum of recommended paths in each recommended path group; recording the ratio of the number to the sum as the segmented trajectory risk of the segmented trajectory; sequentially calculating the segmented trajectory risk of each segmented trajectory, and calculating the average value of each segmented trajectory risk; recording the average value as the trajectory risk of the driving trajectory; generating a detour alarm when the trajectory risk is greater than a set trajectory risk threshold; the trajectory risk threshold is set empirically, and in this embodiment, the trajectory risk threshold is set to 0.4; the detour alarm is to remind passengers that the current route traveled by the driver has detoured. Trip risk detection includes: obtaining the trajectory length of each segment trajectory, summing the trajectory lengths of each trajectory to obtain the distance traveled, and obtaining the recommended update node of the last segment trajectory terminal; obtaining several stage recommended path groups corresponding to the recommended update node, selecting the longest length among the recommended paths in each stage within each recommended path group, and recording it as the distance not yet traveled; when the sum of the traveled distance and the distance not yet traveled is greater than a distance threshold of a set multiple; the set multiple is a number greater than 1 set by experts based on experience, and in this embodiment the set multiple is 1.2 times; generating a detour alarm; It also includes detour risk prediction: obtaining the node detour risk corresponding to the starting path recommendation update point of the segmented trajectory; calculating the average detour risk of each node, and when the average value is greater than the set node risk threshold, the node risk threshold is set by expert experience; generating a detour warning, indicating that the driver is currently driving on a road section with a high detour risk, and the possibility of taking a detour is greater.

[0039] Secondly, this application provides a vehicle detour identification and detection device based on trajectory similarity, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is configured to execute the instructions to implement the method described in the first aspect and any possible implementation thereof. This vehicle detour identification and detection device based on trajectory similarity can be an electronic device or a chip within an electronic device.

[0040] Please see Figure 2 Thirdly, this application provides a vehicle detour identification and detection system based on trajectory similarity, comprising: a data acquisition module, a data analysis module, and a risk warning module; wherein, The data acquisition module includes a network acquisition unit and a vehicle terminal acquisition unit: The network acquisition unit is used to acquire traffic network maps, starting positions, and target positions; The vehicle terminal acquisition unit is used to collect vehicle driving data and driving trajectory during driving; The data analysis module includes a path recommendation unit and a risk assessment unit; The route recommendation unit is used to determine route recommendation update points based on the traffic network map; generate several recommended route groups based on the starting position and the target position; and generate several stage recommended route groups based on driving data and route recommendation update points. The risk assessment unit is used to segment the journey based on the path recommendation update points to obtain several segmented trajectories; and to perform detour risk detection based on the several segmented trajectories and the recommended path groups at each stage; the detour risk detection includes trajectory risk detection and journey risk detection. The risk warning module is used to obtain detour alerts and detour warnings, and to issue alarms.

[0041] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a trajectory similarity-based vehicle detour identification and detection device, cause the trajectory similarity-based vehicle detour identification and detection device to perform the method described in the first aspect and any possible implementation thereof.

[0042] Fifthly, this application provides a computer program product containing instructions that, when the computer program product is run on a vehicle detour identification and detection device based on trajectory similarity, causes the vehicle detour identification and detection device based on trajectory similarity to perform the method described in the first aspect and any possible implementation thereof.

[0043] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0044] How this application works: By acquiring a traffic network map, route recommendation update points are determined based on the map; the starting and destination positions are obtained; several recommended route groups are generated based on the starting and destination positions; vehicle driving data is acquired in real time, and several stage recommended route groups are generated based on the driving data and route recommendation update points; the driving trajectory is acquired, and the driving is segmented based on the route recommendation update points to obtain several segmented trajectories; detour risk detection is performed based on several segmented trajectories and recommended route groups at each stage; the detour risk detection includes trajectory risk detection and trip risk detection; through matching degree analysis between segmented trajectories and recommended route groups, multi-segment detour risk analysis is achieved, increasing the accuracy of detour analysis.

[0045] The above embodiments are only used to illustrate the technical methods of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of this application without departing from the spirit and scope of the technical methods of this application.

Claims

1. A vehicle detour identification and detection method based on trajectory similarity, characterized in that, include: Obtain a traffic network map and determine the route recommendation update point based on the traffic network map; Obtain the starting and target positions; generate several recommended path groups based on the starting and target positions; Real-time acquisition of vehicle driving data, and generation of several stages of recommended path groups based on driving data and path recommendation update points; The driving trajectory is obtained, and the driving is segmented based on the path recommendation update points to obtain several segmented trajectories; detour risk detection is performed based on several segmented trajectories and recommended path groups at each stage; the detour risk detection includes trajectory risk detection and trip risk detection.

2. The vehicle detour identification and detection method based on trajectory similarity according to claim 1, characterized in that, The method of determining route recommendation update points based on traffic network maps includes: Extract each intersection node from the traffic network map; obtain the corresponding node position of each intersection node, and record the node position as the path recommendation update point. The traffic network map is a road network map composed of various roads, and the intersection node is the intersection of different roads.

3. The vehicle detour identification and detection method based on trajectory similarity according to claim 1, characterized in that, The generation of several recommended path groups based on the starting and destination positions includes: Several feasible paths are generated based on the starting and target positions. Several intersection nodes are obtained from each feasible path. The travel distance and travel time between adjacent intersection nodes are calculated. The travel time is... The travel distance of the feasible path is obtained by summing the travel distances between each adjacent intersection node of the feasible path; the travel time of the feasible path is obtained by summing the travel times between each adjacent intersection node of the feasible path. Obtain distance and time thresholds, and record feasible paths whose travel distance is less than the distance threshold and whose travel time is less than the time threshold as recommended paths; Calculate the overlap ratio between each recommended path. When the overlap ratio is greater than a set overlap ratio threshold, the two recommended paths corresponding to the overlap ratio are grouped into the same recommended path group.

4. The vehicle detour identification and detection method based on trajectory similarity according to claim 3, characterized in that, One method for obtaining the node's travel time includes: Obtain historical data for each time period between each intersection node; extract the average travel time of each vehicle from the historical data, and record the average travel time as the travel time for the corresponding time period between the intersection nodes; Obtain the time period for arriving at the intersection node, and the travel time between the intersection node and the next adjacent intersection node within the time period, and record the travel time as the node travel time between the two intersection nodes; the time period for arriving at the intersection node is the time period corresponding to the start time plus the travel time of each node before the intersection node.

5. The vehicle detour identification and detection method based on trajectory similarity according to claim 1, characterized in that, The generation of several-stage recommended path groups based on driving data and path recommendation update points includes: Obtain the target driving route, which is the recommended route currently being traveled; Retrieve location information and direction of travel from the driving data; determine the next recommended update point on the target driving path based on the location information and direction of travel; Several feasible paths are obtained between the node location of the recommended update point and the target location, and the travel distance and travel time of each feasible path are calculated; the initial distance threshold and initial time threshold of the recommended update point are obtained; the initial distance threshold is the product of the distance threshold adjustment factor and the initial average travel distance, and the initial time threshold is the product of the time threshold adjustment factor and the initial average travel time; based on the initial distance threshold and initial time threshold corresponding to several feasible paths, as well as the travel distance and travel time, the node detour risk of the recommended update point is generated; The distance threshold adjustment factor and time threshold adjustment factor are corrected based on the node detour risk, and the corrected node distance threshold and node time threshold are generated. When the travel distance of a feasible path is less than the node distance threshold and the travel time is less than the node time threshold, the feasible path is recorded as the stage recommended path of the recommended node. Calculate the overlap ratio between recommended paths at each stage. When the overlap ratio is greater than a set overlap ratio threshold, the two recommended paths corresponding to the overlap ratio are grouped into the same stage recommended path group.

6. The vehicle detour identification and detection method based on trajectory similarity according to claim 5, characterized in that, The node detour risk generated based on initial distance and initial time thresholds corresponding to several feasible paths, as well as travel distance and travel time, includes: When the driving distance corresponding to a feasible path is greater than the set initial distance threshold and the driving time is greater than the set driving time threshold, the feasible path is recorded as a detour path. When the travel distance corresponding to a feasible path is less than or equal to the set initial distance threshold, or the travel time is less than or equal to the set initial time threshold, the feasible path is marked as a normal path. Based on the travel distance and time of the detour route and the normal route, and by substituting the initial distance threshold and the initial time threshold into the set detour risk quantification function, the detour risk of the corresponding node is obtained.

7. The vehicle detour identification and detection method based on trajectory similarity according to claim 6, characterized in that, The process of correcting the distance threshold adjustment factor and time threshold adjustment factor based on node detour risk, and generating corrected node distance thresholds and node time thresholds, includes: The node detour risk is denoted as The distance threshold adjustment factor is denoted as The time threshold adjustment factor is denoted as ; Through formula The corrected distance threshold adjustment factor was calculated. ; Through formula The corrected time threshold adjustment factor was calculated. ; The product of the distance threshold adjustment factor and the initial average driving distance is recorded as the corrected node distance threshold. The product of the time threshold adjustment factor and the initial average travel time is denoted as the corrected node time threshold.

8. The vehicle detour identification and detection method based on trajectory similarity according to claim 1, characterized in that, The trajectory data is segmented based on path recommendation update points to obtain several segmented trajectory data; including: Extract the driving trajectory from the trajectory data; obtain several path recommendation update points in the traffic network map; segment the driving trajectory based on the path recommendation update points to obtain several segmented trajectories, wherein the segmented trajectories are the driving trajectories between adjacent recommendation update points.

9. A vehicle detour identification and detection method based on trajectory similarity according to claim 1, characterized in that, The detour risk detection includes trajectory risk detection and travel risk detection; The trajectory risk detection includes: obtaining several stage recommended path groups corresponding to the path recommendation update point at the start of the segmented trajectory; extracting the number of stage recommended paths in the several stage recommended path groups that do not contain the segmented trajectory, and the sum of stage recommended paths in each stage recommended path group; recording the ratio of the number to the sum as the segmented trajectory risk of the segmented trajectory; sequentially calculating the segmented trajectory risk of each segmented trajectory, and calculating the average value of each segmented trajectory risk; recording the average value as the trajectory risk of the driving trajectory; and generating a detour alarm when the trajectory risk is greater than a set trajectory risk threshold. The trip risk detection includes: obtaining the trajectory length of each segment trajectory, summing the trajectory lengths of each trajectory to obtain the distance traveled, and obtaining the recommended update node of the last segment trajectory terminal; obtaining several stage recommended path groups corresponding to the recommended update node, selecting the longest length among the recommended paths in each stage within each recommended path group, and recording it as the distance not traveled; when the sum of the traveled distance and the distance not traveled is greater than a set multiple of a distance threshold; generating a detour alarm.

10. A vehicle detour identification and detection system based on trajectory similarity, used in the application of the vehicle detour identification and detection method based on trajectory similarity as described in any one of claims 1 to 9, characterized in that, include: The module includes a data acquisition module, a data analysis module, and a risk warning module; among which, The data acquisition module includes a network acquisition unit and a vehicle terminal acquisition unit: The network acquisition unit is used to acquire traffic network maps, starting positions, and target positions; The vehicle terminal acquisition unit is used to collect vehicle driving data and driving trajectory during driving; The data analysis module includes a path recommendation unit and a risk assessment unit; The route recommendation unit is used to determine route recommendation update points based on the traffic network map; generate several recommended route groups based on the starting position and the target position; and generate several stage recommended route groups based on driving data and route recommendation update points. The risk assessment unit is used to segment the journey based on the path recommendation update points to obtain several segmented trajectories; and to perform detour risk detection based on the several segmented trajectories and the recommended path groups at each stage; the detour risk detection includes trajectory risk detection and journey risk detection. The risk warning module is used to obtain detour alerts and issue alarms.

Images