CN112927497B

CN112927497B - Floating car identification method, related method and device

Info

Publication number: CN112927497B
Application number: CN202110047491.5A
Authority: CN
Inventors: 王雪松; 王广彬; 左帆
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2021-01-14
Filing date: 2021-01-14
Publication date: 2023-01-17
Anticipated expiration: 2041-01-14
Also published as: CN112927497A

Abstract

The invention discloses a floating car identification method, a related method and a related device. The floating car identification method comprises the following steps: determining dynamic characteristic data generated by the floating vehicle on a running road based on the track data returned by the floating vehicle within a preset time; if the driving road is a section with multiple abnormal driving behaviors, acquiring the type of the abnormal driving behaviors which are easy to occur in the driving road; comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the driving road has abnormal driving behavior; and if so, determining whether the corresponding floating vehicle is the floating vehicle with the abnormal driving behavior or not based on the track data of the floating vehicle on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type. The floating car of the abnormal driving behavior is recognized, and the recognition accuracy of the floating car is high.

Description

Floating car identification method, related method and device

Technical Field

The invention relates to the technical field of dynamic traffic, in particular to a floating car identification method, a related method and a related device.

Background

In the prior art, determination of road conditions (smooth running, slow running, congestion and the like) needs to utilize track data returned by floating cars, and if the floating cars participating in the determination of the road conditions have abnormal running behaviors (such as low-speed running or parking caused by non-congestion), the problem of inaccurate road condition determination is caused. Therefore, the method is very important for accurately determining the road condition by identifying whether the abnormal driving behaviors exist in the floating car. Accordingly, it is highly desirable for those skilled in the art to provide a corresponding technical solution that can identify a floating car having the aforementioned abnormal driving behavior.

Disclosure of Invention

In view of the above, the present invention has been developed to provide a floating car identification method, related method and apparatus that overcomes, or at least partially addresses, the above-identified problems.

In a first aspect, an embodiment of the present invention provides a floating car identification method, including:

determining dynamic characteristic data generated by the floating vehicle on a running road based on the track data returned by the floating vehicle within a preset time;

if the driving road is a road section with multiple abnormal driving behaviors, acquiring the type of the abnormal driving behaviors which are easy to occur in the driving road;

comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road;

and if so, determining whether the corresponding floating vehicle is the floating vehicle with the abnormal driving behavior or not based on the track data of the floating vehicle on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type.

In a second aspect, an embodiment of the present invention provides a method for identifying a road with abnormal driving behavior, including:

training a second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type by using the historical characteristic data of the road included in the training sample set;

when the training recall condition is not met, adjusting the threshold value of the second road characteristic parameter, and returning to the step of continuously executing the training;

when the training recall condition is met, verifying the second road characteristic parameter threshold by using the historical characteristic data of the road in a verification sample set;

when the second road characteristic parameter does not meet the verification condition, adjusting the threshold value of the second road characteristic parameter, and returning to the step of continuously executing the training;

and when the verification condition is met, filtering the roads in the road network by using the second road characteristic parameter threshold value based on the historical characteristic data of each road in the road network, and marking the roads meeting the filtering condition as the abnormal driving behavior multi-occurrence road section of the specified abnormal driving behavior type.

In a third aspect, an embodiment of the present invention provides a method for determining a road traffic speed, including:

the floating car identification method is used for identifying and acquiring whether the floating car on the running road is the floating car with abnormal running behavior or not, and determining the abnormal influence factor of the floating car according to the identification result corresponding to the floating car;

obtaining a speed fusion weight of the floating car according to the abnormal influence factor and the obtained characteristic influence factor in the preset scene;

and determining the passing speed of the running road based on the floating vehicle speed on the running road and the corresponding fusion weight.

In a fourth aspect, an embodiment of the present invention provides a method for determining a real-time traffic status, including:

determining the passing speed of a running road by using the road passing speed determination method;

and determining the real-time road condition of the running road based on the passing speed of the running road and the type of the abnormal running behavior which is easy to occur on the running road.

In a fifth aspect, an embodiment of the present invention provides a floating car identification device, including:

the data determination module is used for determining dynamic characteristic data generated by the floating vehicle on a running road based on the track data returned by the floating vehicle within a preset time length;

the type acquisition module is used for acquiring abnormal driving behavior types which are easy to occur in the driving road if the driving road is a road section with multiple abnormal driving behaviors;

the floating car identification module is used for comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road or not; and if so, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type.

In a sixth aspect, an embodiment of the present invention provides an abnormal driving behavior road recognition apparatus, including:

the training module is used for training a second road characteristic parameter threshold value corresponding to the specified abnormal driving behavior type by using the historical characteristic data of the road in the training sample set;

when the training recall condition is not met, adjusting the threshold value of the second road characteristic parameter, and continuing training;

the verification module is used for verifying the second road characteristic parameter threshold by using the historical characteristic data of the road in the verification sample set when the training recall condition is met;

when the second road characteristic parameter does not meet the verification condition, adjusting the threshold value of the second road characteristic parameter, and continuing training;

and the filtering module is used for filtering the roads in the road network by using the second road characteristic parameter threshold value based on the historical characteristic data of each road in the road network when the verification condition is met, and marking the roads meeting the filtering condition as the abnormal driving behavior multi-occurrence road section of the specified abnormal driving behavior type.

In a seventh aspect, an embodiment of the present invention provides a road traffic speed determining device, including:

the data determining module is used for determining dynamic characteristic data generated by the floating vehicle on a running road based on track data returned by the floating vehicle within a preset time;

the floating car identification module is used for comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road or not; if so, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and a track characteristic parameter threshold value corresponding to the abnormal driving behavior type;

the abnormal influence factor acquisition module is used for determining the abnormal influence factor of the floating car on the running road according to the corresponding recognition result of the floating car;

the fusion weight acquisition module is used for acquiring the speed fusion weight of the floating car according to the abnormal influence factors and the acquired characteristic influence factors in each preset scene;

and the speed determining module is used for determining the passing speed of the running road based on the speed of the floating vehicle on the running road and the corresponding speed fusion weight.

In an eighth aspect, an embodiment of the present invention provides a real-time traffic status determining device, including:

the floating car identification module is used for comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road; if so, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and a track characteristic parameter threshold value corresponding to the abnormal driving behavior type;

the speed determining module is used for determining the passing speed of the running road based on the speed of the floating vehicle on the running road and the corresponding speed fusion weight;

the real-time road condition acquisition module is used for determining the real-time road condition of the running road based on the passing speed of the running road and the type of abnormal running behaviors which are easy to occur in the running road.

In a ninth aspect, an embodiment of the present invention provides a real-time road condition determining system, including: a server and at least one client, wherein:

the server is provided with the real-time road condition determining device and is used for receiving the real-time road condition request sent by the client and sending the determined real-time road condition to the client;

and the client is used for sending the real-time road condition request to the server and receiving the real-time road condition returned by the server.

In a tenth aspect, an embodiment of the present invention provides a service, where the service executes at least one of the above-mentioned floating vehicle track identification method, the above-mentioned abnormal driving behavior road identification method, the above-mentioned road traffic speed determination method, and the above-mentioned real-time road condition determination method.

The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:

the floating car identification method provided by the embodiment of the invention obtains dynamic characteristic data of a road on which a floating car runs based on track data returned by the floating car within a preset time, obtains an abnormal running behavior type of the road on which the floating car runs in a predetermined abnormal running behavior multi-occurrence road section when the road on which the floating car runs is determined to be the road on which the abnormal running behavior is multi-occurrence road section, and determines whether the floating car is the floating car with the abnormal running behavior by using the dynamic characteristic data and a first road characteristic parameter threshold value and a track characteristic parameter threshold value corresponding to the abnormal running behavior type. The accuracy of the abnormal driving behavior identification of the floating car is high, the floating car with the normal driving behavior can be prevented from being identified as the floating car with the abnormal driving behavior by mistake, no human intervention is caused in the abnormal driving behavior identification process of the floating car, and the automation degree is high.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

fig. 1 is a flowchart of a floating car identification method according to an embodiment of the present invention;

FIG. 2 is a schematic illustration of an operating vehicle aggregation scenario;

FIG. 3 is a schematic diagram of a vehicle aggregation scenario including a particular point of interest;

FIG. 4 is a schematic view of a vehicle low speed scenario;

fig. 5 is a flowchart of a method for identifying an abnormal driving behavior road according to an embodiment of the present invention;

fig. 6 is a flowchart of updating road network data including road sections with abnormal driving behaviors according to an embodiment of the present invention;

fig. 7 is a schematic flowchart of a specific implementation of the method for identifying a road with an abnormal driving behavior according to the embodiment of the present invention;

fig. 8 is a flowchart of a method for determining a road passing speed according to an embodiment of the present invention;

fig. 9 is a schematic flowchart of a specific implementation of the method for determining a road passing speed according to the embodiment of the present invention;

fig. 10 is a flowchart of a method for determining a real-time traffic status according to an embodiment of the present invention;

fig. 11 is a schematic structural diagram of a floating car recognition device provided by an embodiment of the invention;

FIG. 12 is a schematic structural diagram of another floating car identification apparatus provided in accordance with an embodiment of the present invention;

fig. 13 is a schematic structural diagram of an abnormal driving behavior road recognition apparatus according to an embodiment of the present invention;

fig. 14 is a schematic structural diagram of a road passing speed determination device according to an embodiment of the present invention;

fig. 15 is a schematic structural diagram of a real-time traffic status determining apparatus according to an embodiment of the present invention;

fig. 16 is a schematic structural diagram of a real-time traffic status determining system according to an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

The following provides a detailed description of specific embodiments of a floating car identification method, a related method and a device provided by the embodiments of the present invention.

Example one

The embodiment of the invention provides a floating car identification method, the flow of which is shown in figure 1 and comprises the following steps:

s101: determining dynamic characteristic data generated by the floating vehicle on a running road based on the track data returned by the floating vehicle within a preset time;

s102: if the driving road is a section with multiple abnormal driving behaviors, acquiring the type of the abnormal driving behaviors which are easy to occur in the driving road;

s103: comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road; if yes, executing step S104, otherwise, executing step S105;

s104: determining whether the corresponding floating car is a floating car with abnormal driving behaviors or not based on the track data of the floating car on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type;

s105: it is impossible to determine whether the floating vehicle is a floating vehicle having an abnormal traveling behavior.

In step S101, based on the track data returned by the floating vehicle within the preset time period, the determined dynamic characteristic data generated by the floating vehicle on the driving road may include: the total number of the floating vehicles, the types of the floating vehicles, the speed of the floating vehicles and the like; the floating vehicles can be divided into operating vehicles and non-operating vehicles. When the type of the floating car is an operating car, the track data of the floating car can come from a taxi or a network taxi appointment client side, and when the type of the floating car is a non-operating car, the track data of the floating car can come from a map application client side. The time length of the preset time length can be selected according to actual conditions, and for example, the time length can be 5 minutes or 10 minutes estimated from the current time forward.

In the embodiment of the present invention, the track data returned by the floating car described in the above step S101 includes a sequence of track points in the process from the departure point to the destination of the floating car. Wherein, above-mentioned track point can be GPS track point or the track point of the floating car that obtains through other modes.

In the step S102, it may be determined whether the driving road is the abnormal driving behavior-prone section, and when the driving road is the abnormal driving behavior-prone section, the type of the abnormal driving behavior that is likely to occur in the driving road may be obtained by:

determining whether the running road is a road section with multiple abnormal running behaviors or not based on a road section with multiple abnormal running behaviors in a road network which is predetermined and an abnormal running behavior type which is easy to occur;

and if so, acquiring the abnormal driving behavior type which is easy to occur in the driving road.

The road network with multiple abnormal driving behaviors and the type of the abnormal driving behaviors which are easy to occur in the embodiment of the invention are obtained by filtering the roads in the road network by using a second road characteristic parameter threshold corresponding to a predetermined specified abnormal driving behavior type based on the historical characteristic data of the roads in the road network.

In an embodiment of the present invention, the abnormal driving behavior types of the road sections with multiple abnormal driving behaviors in the road network may include: operating vehicle aggregation, vehicle aggregation involving specific points of interest (POIs), and types of abnormal driving behavior that may be likely to occur in low speed vehicles or other roads. The road section with frequent abnormal driving behaviors and operating vehicle aggregation is generally a road on which passengers prone to live, such as operating vehicles and the like, for example, referring to a scene shown in fig. 2, passengers of operating vehicles near a mall, near an office building and the like frequently get on and off the road, such as operating vehicles and the like prone to live; the abnormal driving behavior-rich road segment where the vehicle aggregation including the specific point of interest occurs may be a road including a specific POI such as a parking lot, a gas station, a logistics company, for example, in the scene shown in fig. 3, the vehicle aggregation occurs in the road including the POI in the parking lot; the sections with multiple abnormal driving behaviors, in which vehicles at low speed are parked illegally, mainly include roads with few road lanes, roads with uneven road surfaces, roads with many pedestrians or non-motor vehicles in the roads, and roads with parks at two sides of the roads, for example, in the scene shown in fig. 4, vehicles at two sides of the roads are parked illegally, which results in low speed of the vehicles.

In this embodiment of the present invention, the described historical feature data of the roads in the road network may include: historical raw data of the road are obtained from road network data, and historical dynamic data of the road are determined based on the obtained track data of a plurality of floating cars matched to the road in a preset time period. The historical raw data of the road may include: at least one of road grade, road type, area where the road is located, traffic light information, toll station information and interest point information of the road; the historical dynamic data of the road may include: at least one of a daily average traffic flow of the road, an average speed of the floating car, an average transit time of the floating car, an average parking time of the floating car, and a type of the floating car.

In step S103, the first road characteristic parameter threshold corresponding to the abnormal driving behavior type may be predetermined, and each abnormal driving behavior type has its corresponding first road characteristic parameter threshold. And then, after determining dynamic characteristic data generated by the floating vehicle on the driving road based on the track data returned by the floating vehicle within a preset time, comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the type of abnormal driving behaviors which are easy to occur in the driving road, so as to determine whether the abnormal driving behaviors occur on the driving road.

In step S104, the track characteristic parameter threshold corresponding to the abnormal driving behavior type may be predetermined, and each abnormal driving behavior type may have a track characteristic parameter threshold corresponding thereto. And then, by comparing the acquired track data returned by the floating vehicles on the running road with the track characteristic parameter threshold value corresponding to the abnormal running behavior type which is easy to occur in the running road, whether the corresponding floating vehicles are the floating vehicles with the abnormal running behaviors can be determined.

In one embodiment, referring to fig. 5, the process of training the second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type may include the following steps:

s201: training a second road characteristic parameter threshold value corresponding to the specified abnormal driving behavior type by using the historical characteristic data of the road in the training sample set; the training sample set comprises a plurality of road samples of abnormal driving behavior multi-road sections of specified abnormal driving behavior types;

s202: judging whether the training sample set meets a training recall condition, if not, executing a step S203, and if so, executing a step S204;

s203: adjusting the second road characteristic parameter threshold value, and returning to continue to execute the step S201;

s204: verifying the second road characteristic parameter threshold value by using the historical characteristic data of the road included in the verification sample set; the verification sample set comprises a plurality of road samples of abnormal driving behavior multi-sending road sections of specified abnormal driving behavior types, and the number of the road samples in the verification sample set is greater than that of the road samples in the training sample set;

s205: judging whether the verification sample set meets the verification condition, if not, executing step S206, and if so, executing step S207;

s206: adjusting the threshold value of the second road characteristic parameter, and returning to continue executing the step S201;

s207: and obtaining a second road characteristic parameter threshold value corresponding to the trained specified abnormal driving behavior type.

In the embodiment of the present invention, according to the steps S201 to S207, the second road characteristic parameter threshold corresponding to the trained specified abnormal driving behavior type is obtained. The specified abnormal traveling behavior type may be one of an operating vehicle cluster, a vehicle cluster containing a specific point of interest, and an abnormal traveling behavior type that is likely to occur in a low-speed or other road of the vehicle.

In a specific embodiment, the training of the second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type in step S201 by using the historical characteristic data of the road included in the training sample set includes:

according to the received second road characteristic parameter threshold in training, screening to obtain road samples meeting the second road characteristic parameter threshold in the training sample set based on historical original data of roads and historical dynamic data of roads of all road samples in the training sample set;

in an embodiment, the step S202 of determining whether the training sample set meets the training recall condition includes:

and judging whether the road samples meeting the second road characteristic parameter threshold value in the training sample set meet a training recall condition or not.

In the embodiment of the present invention, the training recall condition may include: at least one of recall rate, accuracy rate, and accidental injury rate. The implementation of determining whether the training recall condition is met may be performed in the manner known in the art. For example, when the training recall condition is a recall rate, a training recall condition with a met recall rate may be preset, for example, the preset recall rate threshold is 80%, the recall rate during training is determined according to the ratio of the recalled road samples meeting the second road characteristic parameter threshold to all the samples in the training sample set, and is compared with the preset recall rate threshold, and if the recall rate during training is greater than the preset recall rate threshold, the training recall condition is met.

In a specific embodiment, the verifying the second road characteristic parameter threshold by using the historical characteristic data of the road included in the verification sample set in step S204 includes:

and screening to obtain the road sample meeting the second road characteristic parameter threshold value in the verification sample set based on the historical original data of the road and the historical dynamic data of the road sample in the verification sample set according to the second road characteristic parameter threshold value.

In an embodiment, the determining whether the verification sample set meets the verification condition in step S205 includes:

and judging whether the road samples meeting the second road characteristic parameter threshold in the verification sample set meet verification conditions or not.

In the embodiment of the present invention, the verification condition may include: at least one of recall rate, accuracy rate, and accidental injury rate. The implementation process of determining whether the verification condition is satisfied may be implemented in a manner in the prior art. For example, when the verification condition is an accuracy rate, the verification condition with the accuracy rate reaching the standard may be preset, for example, the preset accuracy rate threshold is 90%, then according to the number of the recalled road samples meeting the second road characteristic parameter threshold and the abnormal driving behavior types of the road samples labeled in advance in the verification sample set, the percentage of the number in all the recalled road samples meeting the second road characteristic parameter threshold is determined to obtain the accuracy rate at the time of verification, the accuracy rate at the time of verification is compared with the preset accuracy rate threshold by 90%, and if the accuracy rate at the time of verification is greater than the preset verification accuracy rate threshold, the verification condition is met.

In one embodiment, referring to fig. 5, after the second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type is obtained through training, the road network with multiple abnormal driving behaviors and the type of the abnormal driving behavior that is likely to occur may be determined through the following manners:

s208: and filtering the roads in the road network by using the trained second road characteristic parameter threshold based on the historical characteristic data of each road in the road network, and marking the road meeting the filtering condition as an abnormal driving behavior multi-occurrence road section of the specified abnormal driving behavior type.

In the embodiment of the invention, when a plurality of second road characteristic parameter thresholds corresponding to different abnormal driving behavior types are trained, the roads in the road network are respectively filtered by using the second road characteristic parameter threshold corresponding to each trained specified abnormal driving behavior type based on the historical characteristic data of each road in the road network, and the roads meeting the filtering condition are marked as abnormal driving behavior multi-occurrence road sections corresponding to the abnormal driving behavior types to obtain the multi-occurrence road sections of the abnormal driving behaviors in each road in the road network; and the abnormal driving behavior type which is easy to occur in the road when the road is a multi-road section with the abnormal driving behavior.

In the embodiment of the invention, in order to ensure that the abnormal driving behavior multiple road sections marked by the roads in the road network are real roads easy to generate abnormal driving behaviors, the roads marked as the abnormal driving behavior multiple road sections and the abnormal driving behavior types easy to generate in the road network can be sampled and evaluated, namely, a preset number of the roads marked as the abnormal driving behavior multiple road sections are extracted from the road network, whether the roads are the real abnormal driving behavior multiple road sections or not is judged according to the track data of a plurality of floating cars driving on the roads, whether the real abnormal driving behavior types easy to generate in the roads are consistent with the marked abnormal driving behavior types or not is judged if the roads are the real abnormal driving behavior multiple road sections, and if the road types are the real abnormal driving behavior multiple road sections, the evaluation result of the roads marked as the abnormal driving behavior multiple road sections in the road network is determined to be qualified; and if the ratio of the number of the roads with qualified evaluation results to the preset number is larger than a preset parameter value, for example 80%, from the preset number of the roads marked as the road sections with the abnormal driving behaviors, the roads are extracted from the road network, and the sampling evaluation is qualified.

In the embodiment of the present invention, since the road characteristic defining conditions satisfied by the abnormal driving behavior multi-occurrence road segment of each abnormal driving behavior type are different, that is, each abnormal driving behavior type corresponds to a different second road characteristic parameter threshold, it is necessary to train the second road characteristic parameter threshold corresponding to each abnormal driving behavior type. When training starts, an initial second road characteristic parameter threshold value may be preset according to the historical characteristic data of each road in the training sample set. In the following, by way of example, how to preset the initial second road characteristic parameter threshold of the operation vehicle aggregation, the vehicle aggregation including the specific interest point, and the section where three abnormal driving behaviors of the vehicle are frequent at low speed in the embodiment of the present invention is described.

When training the second road characteristic parameter threshold of the abnormal driving behavior multiple road section gathered by the operating vehicles, determining that the second road characteristic parameter threshold of the abnormal driving behavior multiple road section gathered by the operating vehicles at the initial time is as follows:

the restricted road is a non-high speed, express road and a parking-available urban road; the method comprises the steps of setting a parameter value of daily average flow of a road, setting a parameter value of daily average parking quantity in the road, setting a parameter value of a ratio of the quantity of vehicles in starting to park in the road on an average day to the quantity of parked vehicles, setting a parameter value of a ratio of the quantity of operating vehicles parked in the road to all parked vehicles, and setting a parameter value of a ratio of the daily average quantity of operating vehicles in the road to the daily average flow of the road.

In the embodiment of the invention, because the high-speed roads or the express roads in the road network are generally not allowed to stop, the operating vehicles are not easy to gather, but the operating vehicles of non-urban road have less stop and are not easy to gather, so when the second road characteristic parameter threshold value of the road section with the abnormal driving behaviors and the initial operating vehicle gather is set, the road can be limited, so that the high-speed roads or the express roads in the road network are excluded and the road can be limited to be the urban road capable of stopping; because vehicles are frequently parked on the road aggregated by the operating vehicles, the number of vehicles passing through the road cannot be too large, the number of vehicles parked in the road cannot be too small, the proportion of the number of operating vehicles in all parked vehicles in the road cannot be too small, and the proportion of operating vehicles passing through the road every day in all passing vehicles on the road cannot be too low, so that the parameter value of the daily average flow of the road, the parameter value of the daily average parking number in the road, the parameter value of the proportion of the number of operating vehicles parked in the road in all parked vehicles and the parameter value of the proportion of the daily average passing operating vehicles in the road to the daily average flow need to be reasonably set, so that when the road in the road network is filtered by using the trained second road characteristic parameter threshold, the number of the roads to be filtered is reduced, and the accuracy of the filtered abnormal driving behavior aggregated by the operating vehicles on the road is ensured to be the frequent road section. In the embodiment of the invention, the daily average flow of the road can be an average value of the number of vehicles passing by the road every day.

When the second road characteristic parameter threshold of the abnormal driving behavior multiple road sections gathered by vehicles including the specific interest point is trained, determining that the initial second road characteristic parameter threshold of the abnormal driving behavior multiple road sections gathered by vehicles including the specific interest point is as follows:

limiting roads to be urban roads which are more than a preset distance away from a traffic light, wherein the roads are road sections containing specific interest points; limiting the road condition of the road to be in a non-unblocked state; the method comprises the steps of setting a parameter value of the daily average flow of a road, setting a parameter value of the ratio of the daily average flow of a vehicle with the traffic speed exceeding a preset high-speed threshold value to the road in the road or setting a parameter value of the ratio of the daily average flow of a vehicle with the traffic speed lower than a preset low-speed threshold value to the road in the road, wherein the preset high-speed threshold value is larger than the preset low-speed threshold value.

In the embodiment of the invention, because the vehicles are close to the traffic lights, for example, the vehicles within 100 meters can stop to wait for the red light, in order to distinguish whether the stopped vehicles stop or stop with abnormal driving behaviors, when setting the threshold value of the second road characteristic parameter of the road section with frequent abnormal driving behaviors gathered by the initial operating vehicles, the roads can be limited so as to exclude the roads with the distance less than the preset distance from the traffic lights and limit the roads to be urban roads; since vehicle aggregation is likely to occur on a road segment including a specific point of interest, for example, a road segment including a point of interest of a parking lot entrance/exit type or a road less than a preset distance from a parking lot, it is also necessary to limit the road segment including the specific point of interest; the road condition of the road with the vehicle aggregation is generally in a congestion or slow running state, so that the road condition of the road can be limited to be in a non-unblocked state; because vehicles are frequently parked on the road with the vehicle aggregation containing the specific interest point, the number of the vehicles passing through the road cannot be too large, the ratio of the number of the vehicles passing through the road to the number of all the vehicles in the road cannot be too large, or the ratio of the number of the vehicles passing through the road at a low speed and being parked in the road to the number of all the vehicles in the road cannot be too small, so that the parameter value of the daily average flow of the road, the parameter value of the ratio of the daily average flow of the vehicles with the passing speed exceeding the preset high speed threshold value in the road and the road or the parameter value of the ratio of the daily average flow of the vehicles with the passing speed lower than the preset low speed threshold value in the road and the road need to be reasonably set, so as to reduce the number of the roads to be filtered when the road in the road network is filtered by using the trained second road characteristic parameter threshold value, and ensure the accuracy of the filtered abnormal driving behavior multi-occurrence road section with the vehicle aggregation containing the specific interest point.

When the second road characteristic parameter threshold of the road section with the low-speed abnormal driving behaviors of the vehicle is trained, determining that the second road characteristic parameter threshold of the road section with the low-speed abnormal driving behaviors of the vehicle at the initial time is as follows:

limiting roads to be urban roads which are far away from traffic lights and exceed a preset distance; and setting a parameter value of the daily average flow of the road, and setting a parameter value of a median of the speed of the normally running vehicle when the daily average flow of the road is smaller than a preset flow value. The median of the speeds of the vehicles which normally run refers to the median of the running speeds of all the vehicles which freely pass through the road in the non-congestion state of the road.

In the embodiment of the invention, because the vehicle is close to the traffic light, for example, the vehicle in the road within 100 meters may wait for the red light or the yellow light to run at a low speed, in order to distinguish whether the vehicle running at the low speed decelerates at the traffic light or runs abnormally to cause the low speed of the vehicle, when the threshold value of the second road characteristic parameter of the road section where the abnormal running behavior at the initial low speed of the vehicle is frequently generated is set, the road can be limited, so that the road with the distance less than the preset distance from the traffic light is excluded, and the road is limited to be an urban road; the method has the advantages that the number of vehicles passing through the road cannot be large due to the fact that the low-speed road vehicles of the vehicles run slowly, the passing speed of the vehicles cannot be high due to the fact that the vehicles passing through the low-speed road of the vehicles do not stop or are unstable in running speed, and therefore the parameter value of the daily average flow of the road and the parameter value of the median of the speed of the vehicles running normally when the daily average flow of the road is smaller than the preset flow value can be set reasonably, the number of the roads to be filtered is reduced when the roads in the road network are filtered by the trained second road characteristic parameter threshold, and the accuracy of the filtered abnormal running behavior section with the low speed of the vehicles is guaranteed.

In the embodiment of the invention, whether each road in the road network is a section with multiple abnormal driving behaviors and the type of the abnormal driving behaviors which are easy to occur in the road is determined based on the historical characteristic data of the roads in the road network, so that the determination condition of identifying whether the floating car on the road is the floating car with the abnormal driving behaviors is used, the characteristic dimension of floating car identification is increased, and the accuracy of the floating car identification on the road is improved.

In the embodiment of the present invention, when the road network data used for identifying the road with the excessive abnormal driving behavior and the type of the abnormal driving behavior that is likely to occur in the road network is the history data, and when the obtained road network with the excessive abnormal driving behavior and the type of the abnormal driving behavior that is likely to occur in the road network is used for floating car identification of the abnormal driving behavior, the road identification information (Link ID) in the road network may be changed, for example, the shape of the road in the road network may be changed, or the shape of the road in the road network may not be changed, but the road is newly divided into one or more new roads.

For example, if the road is broken into two new roads when the shape of the road is not changed, the identification information of the road before the break is updated to the identification information of the two new roads in the updated network data, and if the road before the break in the predetermined road network is the abnormal driving behavior-prone road section, the two updated roads can be determined as the abnormal driving behavior-prone road section by performing the road network road mapping, and at the same time, the type of the abnormal driving behavior that is likely to occur in the two updated roads can be determined.

In the embodiment of the present invention, the described road network and road mapping may be implemented in the following ways: and performing road mapping according to the road coordinate information before updating in the road network data and the road coordinate information after updating in the road network data, thereby realizing the matching of the characteristics of the road before updating to the road after updating.

In the embodiment of the present invention, since the roads in the road network are constantly changing, the road network data including the predetermined sections with excessive driving behaviors in the road network and the types of abnormal driving behaviors that are likely to occur may be subjected to road network and road mapping according to a preset time period, for example, 1 month or 3 months, and the identification information of the roads with excessive driving behaviors in the predetermined road network is replaced by the updated identification information of the roads, so as to obtain the updated sections with excessive driving behaviors in the road network and the types of abnormal driving behaviors that are likely to occur.

In one embodiment, as shown in fig. 6, since the predetermined road network data including the road with the multiple abnormal driving behaviors and the types of the abnormal driving behaviors that are likely to occur in the road network are stored offline, for convenience of description, the predetermined road network data including the road with the multiple abnormal driving behaviors and the types of the abnormal driving behaviors that are likely to occur in the road network is referred to as offline road network data in the following embodiments of the present invention. Therefore, when the offline road network data needs to be updated, the offline road network data may be first subjected to data formatting production, that is, the road network data in a clear text is formatted into invisible binary data, then the offline road network data is transmitted to the on-line computing program in an online service data updating manner, and the computing program completes data loading and road network road mapping to obtain the updated road network data including the abnormal driving behavior multiple road sections and the easy-to-occur abnormal driving behavior types in the road network.

In the embodiment of the invention, as the floating vehicles on the road are dynamically changed, whether the abnormal driving behavior corresponding to the type of the abnormal driving behavior occurs in the current time period of the section with the multiple abnormal driving behaviors also needs to be judged according to the dynamic characteristic data generated by the determined floating vehicles on the driving road based on the track data returned by the floating vehicles within the preset time length.

In the step S103, when determining whether the abnormal driving behavior occurs on the driving road, the first road characteristic parameter threshold corresponding to the abnormal driving behavior type may be determined in advance.

In one embodiment, if the abnormal driving behavior type is the operating vehicle aggregation, the first road characteristic parameter threshold value described in step S103 may include: an operating vehicle occupancy threshold and a low speed operating vehicle occupancy threshold.

Based on this, the comparing the dynamic characteristic data with the first road characteristic parameter threshold corresponding to the abnormal driving behavior type in step S103 to determine whether the abnormal driving behavior occurs on the driving road may specifically include:

determining a first proportion of the number of floating cars of the operating vehicles to the total number of the floating cars according to the types of the floating cars in the dynamic characteristic data, and judging whether the first proportion is larger than a threshold value of the proportion of the operating vehicles;

determining the type of the floating car as a second proportion of the floating car with the speed smaller than a first preset speed threshold value in the floating cars of the operating cars according to the type and the speed of the floating car in the dynamic characteristic data, and judging whether the second proportion is larger than the proportion threshold value of the low-speed operating cars or not;

if yes, determining that abnormal driving behaviors occur on the driving road.

The operation vehicle proportion threshold value can be used for representing the threshold value of the proportion of the number of the floating vehicles of which the types are the operation vehicles in the preset duration to the total number of the floating vehicles. The low-speed operation vehicle proportion threshold value can be used for representing a proportion threshold value of the number of floating vehicles of which the types are operation vehicles and the speed is less than a first preset speed threshold value in the preset time length to the number of floating vehicles of which the types are operation vehicles in the preset time length. The size of the first preset speed threshold may be determined by referring to actual conditions of roads in a road network, which may not be specifically limited in the embodiment of the present invention.

In the road section with multiple abnormal running behaviors gathered by the operating vehicles, the number of the operating vehicles is generally larger than that of non-operating vehicles, and the operating vehicles have the phenomena of lying down and the like, so that the speed of the operating vehicles is generally lower or in a parking state, and therefore, the running road with the dynamic characteristic data according with the operating vehicle proportion threshold and the low-speed operating vehicle proportion threshold can be determined to have the abnormal running behaviors by limiting the operating vehicle proportion threshold and the low-speed operating vehicle proportion threshold.

In one embodiment, if the abnormal driving behavior type is an operating vehicle aggregation, the trajectory characteristic parameter threshold described in step S104 may include: a first preset speed threshold.

Based on this, if it is determined that the abnormal driving behavior occurs on the driving road in step S103, the determining step S104 may be implemented to determine whether the corresponding floating vehicle is a floating vehicle with the abnormal driving behavior based on the track data of the floating vehicle on the driving road and the track characteristic parameter threshold corresponding to the abnormal driving behavior type, where the determining step may specifically include:

when the floating car is determined to be an operating car according to the type of the floating car in the track data of the floating car, judging whether the speed of the floating car in the track data of the floating car is smaller than the first preset speed threshold value;

and if so, determining that the floating vehicle is the floating vehicle with abnormal driving behaviors.

In one embodiment, if the abnormal driving behavior type is a vehicle cluster including a specific point of interest, the first road characteristic parameter threshold described in step S103 may include: a first low speed vehicle occupancy threshold;

determining a third ratio of the floating vehicles with the speed smaller than a second preset speed threshold value to the total number of the floating vehicles according to the speed of the floating vehicles in the dynamic characteristic data, and judging whether the third ratio is larger than the first low-speed vehicle ratio threshold value or not;

and if so, determining that abnormal driving behaviors occur on the driving road.

The first low-speed vehicle proportion threshold value can be used for representing a proportion threshold value of the number of the floating vehicles with the speed smaller than a second preset speed threshold value in a preset time period to the total number of the floating vehicles. The size of the second preset speed threshold may be determined by referring to actual conditions of roads in the road network, which may not be specifically limited in the embodiment of the present invention.

In the abnormal driving behavior-rich road section containing the vehicle aggregation of the specific interest point, vehicles are easy to aggregate because vehicles on the road containing the specific interest point avoid vehicles in other directions or have vehicle steering, and the speed of the vehicles at the positions is generally low, so that a first low-speed vehicle occupancy threshold value can be defined, and the driving road with the dynamic characteristic data meeting the first low-speed operation vehicle occupancy threshold value is determined to have abnormal driving behavior.

In one embodiment, if the abnormal driving behavior type is a vehicle cluster including a specific point of interest, the track characteristic parameter threshold described in step S104 may include: a second preset speed threshold.

judging whether the speed of the floating car in the track data of the floating car is smaller than the second preset speed threshold value or not;

In one embodiment, if the abnormal driving behavior type is a low speed of the vehicle, the first road characteristic parameter threshold described in the step S103 may include: a total number threshold of floating vehicles and a second low-speed vehicle occupancy threshold.

judging whether the total number of the floating cars in the dynamic characteristic data is smaller than a threshold value of the total number of the floating cars;

determining a fourth proportion of floating cars with the speed less than a third preset speed threshold value to the total number of the floating cars according to the speed of the floating cars in the dynamic characteristic data, and judging whether the fourth proportion is greater than a second low-speed vehicle proportion threshold value or not;

if yes, determining that abnormal driving behaviors occur on the driving road.

The total number threshold of the floating cars can be used for representing an upper limit threshold of the total number of the floating cars in a preset duration. The second low-speed vehicle proportion threshold value can be used for representing a proportion threshold value of the number of the floating vehicles with the speed smaller than a third preset speed threshold value in the preset duration to the total number of the floating vehicles. The size of the third preset speed threshold may be determined by referring to actual conditions of roads in the road network, which may not be specifically limited in the embodiment of the present invention.

In a road section with a plurality of abnormal driving behaviors, the speed of a vehicle normally driving on a road is much lower than that of a vehicle on a common road, the number of vehicles on the road is generally small, and the proportion of low-speed vehicles in floating vehicles within a preset time is generally large, for example, 80% or 90% in floating vehicles which cannot pass through the road quickly. Therefore, a threshold value of the total number of floating vehicles and a threshold value of the second low-speed vehicle ratio can be defined, and abnormal driving behaviors of the driving road with the dynamic characteristic data meeting the threshold value of the total number of floating vehicles and the threshold value of the second low-speed vehicle ratio are determined.

In one embodiment, if the abnormal driving behavior type is a low speed of the vehicle, the track characteristic parameter threshold described in step S104 may include: a third preset speed threshold;

judging whether the speed of the floating car in the track data of the floating car is smaller than a third preset speed threshold value or not;

The floating car identification method provided by the embodiment of the invention obtains dynamic characteristic data of a road on which a floating car runs based on track data returned by the floating car within a preset time, obtains an abnormal running behavior type of the road on which the floating car runs in a predetermined abnormal running behavior multi-occurrence road section when the road on which the floating car runs is determined to be the road on which the abnormal running behavior is multi-occurrence road section, and determines whether the floating car is the floating car with the abnormal running behavior by using the dynamic characteristic data and a first road characteristic parameter threshold value and a track characteristic parameter threshold value corresponding to the abnormal running behavior type. The accuracy of the abnormal running behavior recognition of the floating car is high, the floating car with the normal running behavior can be prevented from being recognized as the floating car with the abnormal running behavior by mistake, no human intervention is caused in the abnormal running behavior recognition process of the floating car, and the automation degree is high.

Example two

Based on the same inventive concept, referring to fig. 5, an embodiment of the present invention further provides a method for identifying an abnormal driving behavior road, including:

s203: adjusting the threshold value of the second road characteristic parameter, and returning to continue executing the step S201;

s204: verifying the second road characteristic parameter threshold value by using the historical characteristic data of the road included in the verification sample set; the verification sample set comprises a plurality of road samples of abnormal driving behavior multi-road sections of specified abnormal driving behavior types, and the number of the road samples in the verification sample set is larger than that of the road samples in the training sample set;

s205: judging whether the verification sample set meets a verification condition, if not, executing a step S206, and if so, executing a step S207;

s206: adjusting the second road characteristic parameter threshold value, and returning to continue to execute the step S201;

s207: obtaining a second road characteristic parameter threshold value corresponding to the trained specified abnormal driving behavior type;

In a specific embodiment, referring to fig. 7, before training a second road characteristic parameter threshold corresponding to a specified abnormal driving behavior type, a training sample set is obtained, and according to historical characteristic data of a road sample of a multi-road-section abnormal driving behavior of each specified abnormal driving behavior type in the training sample set, parameter values of each road characteristic and road characteristic are determined to obtain a second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type at the initial time; training a second road characteristic parameter threshold corresponding to the initial specified abnormal driving behavior type, judging whether a training sample set meets a training recall condition, if not, adjusting the second road characteristic parameter threshold, returning to the step of continuously executing the training, and verifying the second road characteristic parameter threshold in the training by using the historical characteristic data of the road included in the verification sample set until the training recall condition is met; if the training condition is not met, adjusting a second road characteristic parameter threshold value in the training, returning to the step of continuously executing the training until the training condition is met, and obtaining a second road characteristic parameter threshold value corresponding to the trained specified abnormal driving behavior type;

and filtering the roads in the road network by using the trained second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type based on the historical characteristic data of each road in the road network, and marking the road meeting the filtering condition as the abnormal driving behavior multi-occurrence road section of the specified abnormal driving behavior type.

Further, in order to ensure that the abnormal driving behavior sections marked by the roads in the road network are real roads prone to abnormal driving behaviors, sampling evaluation can be performed on the roads marked as the abnormal driving behavior sections and the types of the abnormal driving behaviors prone to occur in the road network, and if the ratio of the number of the roads with qualified evaluation results to the preset number of the roads marked as the abnormal driving behavior sections extracted from the road network is larger than a preset parameter value, for example 80%, the sampling evaluation is determined to be qualified. And determining the road with the abnormal driving behaviors which is obtained by filtering the second road characteristic parameter threshold corresponding to the trained specified abnormal driving behavior type as a real road easy to generate the abnormal driving behaviors.

The road identification method provided by the embodiment of the invention determines whether each road in the road network is a road section with multiple abnormal driving behaviors and the type of the abnormal driving behaviors which are easy to occur on the road based on the historical characteristic data of the roads in the road network, so as to serve as a judgment condition for identifying whether a floating car which runs on the road is a floating car with the abnormal driving behaviors, and improve the accuracy of the identification of the floating car on the road.

EXAMPLE III

Based on the same inventive concept, an embodiment of the present invention further provides a method for determining a road traffic speed, which is shown in fig. 8 and includes:

executing the floating car identification method described in the steps S101 to S105, and identifying whether the floating car on the driving road is a floating car with abnormal driving behavior;

s106: determining an abnormal influence factor of the floating car according to the corresponding recognition result of the floating car;

s107: obtaining the speed fusion weight of the floating car according to the abnormal influence factors and the obtained characteristic influence factors in each preset scene;

s108: and determining the passing speed of the running road based on the floating vehicle speed on the running road and the corresponding fusion weight.

In the third embodiment of the present invention, the specific implementation process of the floating car identification method described in the foregoing steps S101 to S105 can refer to the detailed description in the first embodiment, and is not described again here.

In an embodiment of the present invention, the determining the abnormal influence factor of the floating car according to the identification result corresponding to the floating car in step S105 includes:

if the identified floating car is the floating car with abnormal driving behavior, determining the abnormal influence factor of the floating car on the road as the weight-reducing influence factor;

if the identified floating car is a floating car without abnormal driving behaviors, determining the abnormal influence factor of the floating car on the road as an ascending influence factor;

if the floating vehicle cannot be identified as the floating vehicle with abnormal driving behaviors, determining that the abnormal influence factor of the floating vehicle on the road is 1; the impact factor of the falling weight is less than 1 and the impact factor of the rising weight is greater than 1.

In the embodiment of the invention, because the track data of each floating car in the road are different, the situation that whether the floating car is the floating car with abnormal driving behaviors or not can not be identified may occur on the basis of the track data of the floating car and the track characteristic parameter threshold value. For example, in a road on which an operating vehicle aggregation is likely to occur, if a floating vehicle meets a track characteristic parameter threshold corresponding to the operating vehicle aggregation, it may be determined that the floating vehicle is a floating vehicle with an abnormal driving behavior, and an abnormal influence factor of the floating vehicle on the road on which the floating vehicle is driven is an influence factor of a decreasing right; if the floating vehicle does not meet the track characteristic parameter threshold value corresponding to the running vehicle aggregation, determining that the floating vehicle is a floating vehicle track of normal running behavior, wherein the abnormal influence factor of the floating vehicle on the running road is an ascending influence factor; if the floating vehicle part meets the track characteristic parameter threshold value corresponding to the operation vehicle aggregation, for example, although the type of the floating vehicle is an operation vehicle, if the speed of the floating vehicle is greater than a first preset speed threshold value, that is, the floating vehicle normally passes through the road, it cannot be determined whether the floating vehicle is a floating vehicle with abnormal driving behavior, and at this time, when the passing speed of the road is determined, the abnormal influence factor of the floating vehicle on the road may be not considered, that is, the abnormal influence factor of the floating vehicle on the driving road is determined to be 1.

In the embodiment of the present invention, when determining that a road is a section with multiple abnormal driving behaviors and the type of the abnormal driving behavior that is easy to occur is only one condition of an influence factor that affects the speed of a floating car existing in the road, when determining the traffic speed of the road, it is further required to determine characteristic influence factors of the speeds of floating cars in other preset scenes, which may include:

under the scene of offset between the floating car and the current time, determining a characteristic influence factor related to the time of the floating car according to the time difference between the time when the floating car passes through the running road and the current time, wherein the smaller the time difference between the time when the floating car passes through the running road and the current time is, the more credible the floating car is, and the larger the characteristic influence factor related to the time is.

For example, assuming that the current time is 12 points, wherein the time for the first floating vehicle to pass through the traveled road is 11 points 59 minutes and the time for the second floating vehicle to pass through the traveled road is 11 points 58 minutes in the floating vehicles within the preset time period, the first floating vehicle is closer to the current time, and the characteristic influence factor of the floating vehicle related to the time is larger.

Under the speed distribution scene of the floating car, determining characteristic influence factors related to speed distribution of the floating car according to the section of the speed of the floating car passing through a running road in the speed distribution graph of the floating cars in the preset duration, wherein the closer the speed distribution position of the floating car is to the speed distribution section of the floating car with a larger proportion, the more credible the floating car is, and the larger the characteristic influence factor related to the speed distribution is.

For example, assuming that the number of floating cars on a driving road in a preset time period is 100, wherein the number of floating cars with the speed around 50km/h is 90, and the number of floating cars with the speed around 40km/h is 10, the closer the speed of the floating car is to the speed distribution section of 50km/h, the larger the characteristic influence factor of the floating car related to the speed distribution is.

Under the scene of the acquisition time interval of the floating car, determining characteristic influence factors related to the acquisition time interval of the floating car according to the difference value between the acquisition time interval of each track point in the track data of the floating car and a preset acquisition time interval threshold value, wherein the smaller the difference value between the acquisition time interval of the track points in the track data of the floating car and the preset time interval threshold value is, the more credible the floating car is, and the larger the characteristic influence factor related to the acquisition time interval is.

For example, assuming that a preset collection time interval threshold is 10s, assuming that, in a floating car within a preset time period, a collection time interval of a track point in track data of a first floating car is 5s, and a collection time interval of a track point in track data of a second floating car is 15s, a difference value between the first floating car and the preset collection time interval threshold is a negative value, a characteristic influence factor related to the collection time interval of the floating car is larger, a difference value between the second floating car and the preset collection time interval threshold is a positive value, and a characteristic influence factor related to the collection time interval of the floating car is smaller.

Under the scene of the coverage range of the floating car on the road, according to the track data of the floating car, the track point sequence of the floating car is matched with the road to obtain the coverage range of the track point sequence of the floating car on the road, and the characteristic influence factor related to the coverage range of the floating car and the road is determined.

For example, in a floating car within a preset time, if a track point in a track point sequence of a first floating car covers the whole distance of a road and a track point in a track point sequence of a second floating car covers half of the distance length of the road, the coverage range of the track point sequence of the first floating car on the road is larger, the characteristic influence factor of the floating car related to the coverage range on the road is larger, the coverage range of the track point sequence of the second floating car on the road is smaller, and the characteristic influence factor of the floating car related to the coverage range on the road is smaller.

Assuming that the number of abnormality influencing factors and other characteristic influencing factors is n for each floating car; k for influencing factor _j Represents, wherein j =1, \8230;, n; the speed fusion weight of the floating car is represented by w; then, referring to equation (1), the velocity fusion weight w of the floating car is:

in the formula (1)

Assuming that the number of floating cars in a preset time duration is m, the speed of the floating cars is represented by Vi, wherein i =1, \8230;, m; when the road traffic speed is represented by V, the road traffic speed V is represented by the following equation (2):

V=

in the formula (2)

In the embodiment of the invention, as the floating vehicles running on the road are changed in real time, in order to obtain the traffic speeds of the roads at different time, the traffic speed of the road can be periodically calculated according to a certain time interval, for example, the floating vehicles within a preset time length are obtained every 1 minute or 2 minutes, and the method for determining the traffic speed of the road is executed to obtain the traffic speed of the road.

In a specific embodiment, referring to fig. 9, when determining the traffic speed of a road, it is necessary to determine an obtained abnormal influence factor of a floating car within a preset time period and a feature influence factor in a preset scene, where the determination of the feature influence factor and the abnormal influence factor in the preset scene may be performed simultaneously or step by step;

when determining the characteristic influence factors of the floating car in different preset scenes, the description about the characteristic influence factors of the speed of the floating car in the preset scenes can be referred to;

when determining the abnormal influence factor of the floating car, determining whether the floating car is a floating car with abnormal driving behavior according to the floating car identification method described in the steps S101 to S105, if so, determining that the abnormal influence factor of the floating car is an influence factor of weight reduction, if not, determining that the abnormal influence factor of the floating car is an influence factor of weight increase, and if not, determining that the abnormal influence factor of the floating car is 1;

referring to fig. 9, if it is determined that the road on which the floating car is traveling is in an abnormal traveling behavior, when it is determined that the floating car has an abnormal traveling behavior, it may be determined whether the floating car is a normal floating car, and if it is determined that the floating car is not a normal floating car, it is determined whether the floating car is a floating car having an abnormal traveling behavior, and if it is determined that the floating car is not a floating car having an abnormal traveling behavior, it is determined that it is not possible to determine whether the floating car is a floating car having an abnormal traveling behavior; when it is determined that the road is not a road on which the abnormal driving behavior occurs, it is also considered that it is impossible to determine whether the floating vehicle is a floating vehicle having the abnormal driving behavior.

Then, according to the obtained abnormal influence factor of the floating car and the characteristic influence factor in the preset scene, the speed fusion weight of the floating car is obtained by applying the formula (1);

and finally, obtaining the passing speed of the road by applying the formula (2) according to the obtained speed and speed fusion weight of each floating vehicle in the preset time length.

Example four

Based on the same inventive concept, an embodiment of the present invention further provides a method for determining a real-time road condition, as shown in fig. 10, including:

executing the road passing speed determining method described in the steps S101 to S108 to determine the passing speed of the road on which the floating car runs;

s109: and determining the real-time road condition of the running road based on the passing speed of the running road and the type of the abnormal running behavior which is easy to occur in the running road.

In the third embodiment of the present invention, the specific implementation process of the method for determining a road passing speed, which is described in the foregoing steps S101 to S108, may refer to the detailed description in the foregoing first to third embodiments, and is not described herein again.

In an embodiment, the determining the real-time traffic condition of the driving road in step S109 based on the passing speed of the driving road and the type of the abnormal driving behavior that is likely to occur in the driving road may specifically include:

matching the passing speed of the running road with a preset speed threshold value range corresponding to the abnormal running behavior type which is easy to occur in the running road:

if the traffic speed of the running road is less than the minimum speed threshold value of the preset speed threshold value range, determining that the real-time road condition of the running road is congestion;

if the passing speed of the running road is greater than the maximum speed threshold value of the preset speed threshold value range, determining that the real-time road condition of the running road is smooth;

and if the passing speed of the running road is between the minimum speed threshold and the maximum speed threshold, determining that the real-time road condition of the running road is slow running.

In the embodiment of the invention, in order to accurately release the real-time road condition of the road on the road section with the multiple abnormal driving behaviors, the speed threshold range corresponding to the traffic state of the road can be preset aiming at the road section with the multiple abnormal driving behaviors of different abnormal driving behavior types, so that the traffic state of the road is determined to be smooth, slow to run or blocked by comparing the traffic speed of the road with the minimum speed threshold and/or the maximum speed threshold in the preset speed threshold range. Since the speed threshold ranges of the road traffic state in the abnormal driving behavior frequent road sections of different abnormal driving behavior types are different, different speed threshold ranges need to be set for different abnormal driving behavior types.

For example, if the type of abnormal driving behavior that is likely to occur in the road on which the floating car travels is the operating vehicle aggregation, the speed of the vehicle that normally travels on the traveling road is the same as or similar to the speed of the vehicle that normally travels on the ordinary road (i.e., the road on which the abnormal driving behavior is not likely to occur), and therefore, for a section where abnormal driving behavior that is the operating vehicle aggregation is frequently occurring, the corresponding speed threshold range when the road traffic state is clear, slow traveling, or congested may be the same as the speed threshold range of the ordinary road; when the abnormal driving behavior type which is easy to occur in the driving road is the low speed of the vehicle, even if the driving road has few vehicles or no other vehicles, the speed of the floating car which normally drives on the driving road is far lower than that of the floating car which normally drives on the ordinary road, therefore, for the abnormal driving behavior section of which the abnormal driving behavior type is the low speed of the vehicle, the corresponding speed threshold range can be lower than that of the ordinary road when the road is in a smooth, slow or congested state.

In the method for determining real-time road conditions provided in the prior art, user behaviors are analyzed based on user behavior feature data, user tags are set, and then when the road conditions are calculated in real time, whether a user is an abnormal user or not is determined based on the user tags, the abnormal user on the road is filtered out, and the feature data of a normal user on the road is used for calculating road condition information of the road.

In the specific implementation process of analyzing the user behavior and setting the user label based on the user behavior characteristic data, the user can be marked as a user of a certain category according to the user category which is divided in advance. For example, if it is determined that the user a frequently moves at night and the moving range is generally within a fixed urban area range according to the behavior feature data of the user a, it may be indicated that the user a is a taxi driver, and the user a is marked as the taxi driver; if the situation that the user B often walks at a high speed, passes across cities and has a driving track all day is determined according to the behavior characteristic data of the user B, the user B is indicated as a truck driver, and the user B is marked as the truck driver; assuming that it is determined that the vehicle speed of the user C is low (for example, lower than 40 km/h), the moving range is within a preset distance range (for example, 10 km), the user C often stays at the gate of a cell and the gate of a restaurant, the user C has little activity at night, and the user C has little activity during non-meal time, the user C is identified as a takeaway, and the user C is marked as a takeaway.

In the scheme of presetting a user label for a user and distinguishing whether the user is an abnormal user according to the user label in real-time road condition calculation, because the user quantity is huge, user behavior characteristic data is obtained, when the user label is analyzed and set, the calculated and processed data quantity is huge, the processing process is complex, and because the user behavior is changed, if the user behavior characteristic data is changed, the user needs to be marked again, so the process of analyzing and marking the user behavior characteristic data needs to be continuously carried out, and a large amount of resources are consumed.

For example, after the user D is marked as a truck driver, if the user D drives a non-operating vehicle owned by the user D for a period of time to go on vacation, analysis is performed according to user behavior feature data of the user D, and it is necessary to remove an originally marked truck driver tag of the user D and mark a new user tag again. Or, assuming that after the user E is marked as a takeaway, the user E does not work with the takeaway after a period of time, but becomes a taxi driver, at this time, analysis is performed according to the user behavior feature data of the user E, so that the label of the takeaway marked originally by the user E needs to be removed, and a new user label marked again is a taxi driver.

When the real-time road condition of the road is determined, the number of users driving on the road is limited, and although a large amount of data of the users are loaded in the real-time road condition processing process, most of the users may not travel in the current time period and cannot be used for calculating the real-time road condition of the road when the real-time road condition of the road is actually calculated, so that the cache resources and the calculation resources are wasted due to the fact that a large amount of data of the users without traveling are loaded.

Compared with the method for analyzing the user behaviors, setting the user labels and calculating the road conditions in real time based on the user behavior characteristic data, the method for determining the real-time road conditions determines whether the road is a road easy to generate abnormal driving behaviors and an abnormal driving behavior type easy to generate by using the historical data characteristics of the road, and because the historical data characteristics of the road do not change greatly due to the number of the obtained floating cars, the historical data characteristics of the road occupy less data resources, and whether the road in a finally determined road network is a multi-section with abnormal driving behaviors and occupies less data resources, the amount of loaded data resources in the process of determining the real-time road conditions is small, and the waste of cache resources and calculation resources is avoided; meanwhile, according to the multiple sections of abnormal driving behaviors and the types of abnormal driving behaviors which are easy to occur in the road network, whether the road on which the floating car runs is the multiple sections of abnormal driving behaviors and the types of abnormal driving behaviors which are easy to occur is determined, and the real-time road condition of the road is determined by combining the track data returned by the floating car within the preset time length which is obtained in real time.

Based on the same conception, the embodiment of the invention also provides a floating car identification device, an abnormal driving behavior road identification device, a road traffic speed determination device, a real-time road condition determination device and services.

An embodiment of the present invention further provides a floating car identification device, and as shown in fig. 11, the device includes:

the data determining module 101 is used for determining dynamic characteristic data generated by the floating vehicle on a running road based on track data returned by the floating vehicle within a preset time length;

the type obtaining module 102 is configured to obtain a type of an abnormal driving behavior that is likely to occur in the driving road if the driving road is a road section where abnormal driving behaviors are frequently generated;

the floating car identification module 103 is configured to compare the dynamic characteristic data with a first road characteristic parameter threshold corresponding to the abnormal driving behavior type, and determine whether an abnormal driving behavior occurs on the driving road; and if so, determining whether the corresponding floating vehicle is the floating vehicle with the abnormal driving behavior or not based on the track data of the floating vehicle on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type.

In one embodiment, if the abnormal driving behavior type is an operating vehicle aggregation, the first road characteristic parameter threshold includes: an operating vehicle proportion threshold and a low-speed operating vehicle proportion threshold; the track characteristic parameter threshold comprises: the first preset speed threshold;

the floating car identification module 103 is specifically configured to determine, according to the type of a floating car in the dynamic characteristic data, a first proportion of the number of floating cars of an operating car to the total number of floating cars, and determine whether the first proportion is greater than the operating car proportion threshold; according to the type and the speed of the floating car in the dynamic characteristic data, determining that the type of the floating car is a second proportion of the floating car with the speed smaller than a first preset speed threshold value in the floating car of the operating car, and judging whether the second proportion is larger than the low-speed operating car proportion threshold value or not; if yes, determining that abnormal driving behaviors occur on the driving road;

the speed judging unit is used for judging whether the speed of the floating car in the track data of the floating car is less than the first preset speed threshold value or not when the floating car is determined to be an operating car according to the type of the floating car in the track data of the floating car; and if so, determining that the floating vehicle is the floating vehicle with abnormal driving behaviors.

In one embodiment, if the abnormal driving behavior type is a vehicle aggregation including a specific point of interest, the first road characteristic parameter threshold includes: a first low speed vehicle occupancy threshold; the track characteristic parameter threshold comprises: the second preset speed threshold;

the floating vehicle identification module 103 is specifically configured to determine, according to the speed of the floating vehicle in the dynamic characteristic data, a third proportion of the floating vehicles with a speed smaller than a second preset speed threshold to the total number of the floating vehicles, and determine whether the third proportion is larger than the first low-speed vehicle proportion threshold; if so, determining that abnormal driving behaviors occur on the driving road;

the speed judging module is used for judging whether the speed of the floating car in the track data of the floating car is smaller than a second preset speed threshold value or not; and if so, determining that the floating vehicle is the floating vehicle with abnormal driving behaviors.

In one embodiment, if the abnormal driving behavior type is a low speed of the vehicle, the first road characteristic parameter threshold includes: a total number threshold value of the floating vehicles and a second low-speed vehicle proportion threshold value; the track characteristic parameter threshold comprises: the third preset speed threshold;

the floating car identification module 103 is specifically configured to determine whether the total number of floating cars in the dynamic characteristic data is less than the threshold value of the total number of floating cars; determining a fourth proportion of floating cars with the speed smaller than a third preset speed threshold value to the total number of the floating cars according to the speed of the floating cars in the dynamic characteristic data, and judging whether the fourth proportion is larger than a second low-speed car proportion threshold value or not; if yes, determining that abnormal driving behaviors occur on the driving road;

the speed judging module is used for judging whether the speed of the floating car in the track data of the floating car is smaller than a third preset speed threshold value or not; and if so, determining that the floating car is the floating car with abnormal driving behaviors.

In an embodiment, the type obtaining module 102 is specifically configured to determine whether the driving road is an abnormal driving behavior multi-occurrence road segment based on a predetermined abnormal driving behavior multi-occurrence road segment in a road network and an abnormal driving behavior type that is likely to occur;

In one embodiment, the floating car recognition device, as shown in fig. 12, further includes:

the training module 201 is configured to train a second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type by using the historical characteristic data of the road included in the training sample set;

the verification module 202 is configured to verify the second road characteristic parameter threshold by using historical characteristic data of a road included in a verification sample set when a training recall condition is met;

and the filtering module 203 is configured to, when the verification condition is met, filter the roads in the road network by using the second road characteristic parameter threshold based on the historical characteristic data of each road in the road network, and mark the road meeting the filtering condition as the abnormal driving behavior multi-occurrence road segment of the specified abnormal driving behavior type.

In one embodiment, the floating car recognition device further comprises:

the data acquisition module is used for acquiring historical raw data of a road from road network data, wherein the historical raw data of the road comprises: at least one of road grade, road type, area where the road is located, traffic light information, toll station information and interest point information of the road;

the method comprises the steps of acquiring track data of a plurality of floating cars matched with a road in a preset time period, and determining historical dynamic data of the road; the historical dynamic data of the road comprises: at least one of a daily average traffic flow of the road, an average speed of the floating car, an average transit time of the floating car, an average parking time of the floating car, and a type of the floating car.

In an embodiment, the training module 201 is specifically configured to, according to the received second road characteristic parameter threshold in training, screen and obtain a road sample that satisfies the second road characteristic parameter threshold in the training sample set based on historical raw data of a road and historical dynamic data of a road of each road sample in the training sample set; the training sample set comprises a plurality of road samples of abnormal driving behavior multi-road sections of specified abnormal driving behavior types;

In an embodiment, the verification module 202 is specifically configured to, according to the second road characteristic parameter threshold, obtain, by screening, based on historical raw data of a road and historical dynamic data of the road in a verification sample set, a road sample in the verification sample set that meets the second road characteristic parameter threshold; the verification sample set comprises a plurality of road samples of abnormal driving behavior multi-sending road sections of specified abnormal driving behavior types, and the number of the road samples in the verification sample set is larger than that of the road samples in the training sample set;

Based on the same inventive concept, referring to fig. 13, an embodiment of the present invention further provides an abnormal driving behavior road recognition apparatus, including:

the verification module 202 is configured to verify the second road characteristic parameter threshold by using the historical characteristic data of the road included in the verification sample set when the training recall condition is met;

Based on the same inventive concept, referring to fig. 14, an embodiment of the present invention further provides a road traffic speed determining apparatus, including:

the floating car identification module 103 is configured to compare the dynamic characteristic data with a first road characteristic parameter threshold corresponding to the abnormal driving behavior type, and determine whether an abnormal driving behavior occurs on the driving road; if yes, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type;

an abnormal influence factor obtaining module 104, configured to determine an abnormal influence factor of a floating vehicle on the running road according to a corresponding recognition result of the floating vehicle;

the fusion weight acquisition module 105 is used for obtaining the speed fusion weight of the floating car according to the abnormal influence factors and the obtained characteristic influence factors in each preset scene;

and the speed determining module 106 is used for determining the passing speed of the running road based on the floating vehicle speed on the running road and the corresponding speed fusion weight.

Based on the same inventive concept, referring to fig. 15, an embodiment of the present invention further provides a real-time traffic status determining apparatus, including:

the floating car identification module 103 is configured to compare the dynamic characteristic data with a first road characteristic parameter threshold corresponding to the abnormal driving behavior type, and determine whether an abnormal driving behavior occurs on the driving road; if so, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and a track characteristic parameter threshold value corresponding to the abnormal driving behavior type;

an abnormal influence factor acquiring module 104, configured to determine an abnormal influence factor of a floating car on the running road according to a corresponding recognition result of the floating car;

a speed determination module 106, configured to determine a traffic speed of the driving road based on the floating vehicle speed on the driving road and the corresponding speed fusion weight;

the real-time road condition obtaining module 107 is configured to determine a real-time road condition of the driving road based on a traffic speed of the driving road and a type of an abnormal driving behavior that is likely to occur in the driving road.

Based on the same inventive concept, referring to fig. 16, an embodiment of the present invention further provides a real-time traffic status determining system, including: a server 1 and at least one client 2, wherein:

the server 1 is provided with the real-time road condition determining device, and is used for receiving the real-time road condition request sent by the client 2 and sending the determined real-time road condition to the client 2;

the client 2 is configured to send the real-time road condition request to the server 1, and receive a real-time road condition returned by the server 1.

Based on the same inventive concept, the embodiment of the invention also provides a service, and the service executes at least one of the floating car identification method, the abnormal driving behavior road identification method, the road traffic speed determination method and the real-time road condition determination method during operation.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

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

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims

1. A floating car identification method, comprising:

determining whether the running road is a road section with multiple abnormal running behaviors or not based on a predetermined road section with multiple abnormal running behaviors in a road network and a type of the abnormal running behaviors which are easy to occur;

if so, acquiring the type of abnormal driving behaviors which are easy to occur in the driving road;

and if so, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type.

2. The method according to claim 1, wherein if the abnormal driving behavior type is an aggregation of operating vehicles, the first road characteristic parameter threshold value comprises: an operating vehicle proportion threshold and a low-speed operating vehicle proportion threshold;

the step of comparing the dynamic characteristic data with a first road characteristic parameter threshold corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road comprises the following steps:

according to the type and the speed of the floating car in the dynamic characteristic data, determining that the type of the floating car is a second proportion of the floating car with the speed smaller than a first preset speed threshold value in the floating car of the operating car, and judging whether the second proportion is larger than the low-speed operating car proportion threshold value or not;

3. The method according to claim 2, wherein if the abnormal driving behavior type is an aggregation of operating vehicles, the track characteristic parameter threshold comprises: the first preset speed threshold;

the determining whether the corresponding floating car is the floating car with the abnormal driving behavior based on the track data of the floating car on the driving road and the track characteristic parameter threshold corresponding to the abnormal driving behavior type comprises the following steps:

4. The method according to claim 1, wherein if the abnormal driving behavior type is a vehicle aggregation including a specific point of interest, the first road characteristic parameter threshold value comprises: a first low speed vehicle occupancy threshold;

determining a third proportion of floating vehicles with the speed smaller than a second preset speed threshold value to the total number of the floating vehicles according to the speed of the floating vehicles in the dynamic characteristic data, and judging whether the third proportion is larger than the first low-speed vehicle proportion threshold value or not;

and if so, determining that the abnormal driving behaviors occur on the driving road.

5. The method of claim 4, wherein if the abnormal driving behavior type is a vehicle cluster containing a specific point of interest, the track characteristic parameter threshold comprises: the second preset speed threshold;

6. The method according to claim 1, wherein if the abnormal driving behavior type is a low speed of the vehicle, the first road characteristic parameter threshold value comprises: the total number threshold value of the floating vehicles and the ratio threshold value of the second low-speed vehicles are obtained;

determining a fourth proportion of floating cars with the speed smaller than a third preset speed threshold value to the total number of the floating cars according to the speed of the floating cars in the dynamic characteristic data, and judging whether the fourth proportion is larger than a second low-speed car proportion threshold value or not;

if yes, determining that abnormal driving behaviors occur on the driving road.

7. The method according to claim 6, wherein if the abnormal driving behavior type is low speed of the vehicle, the track characteristic parameter threshold value comprises: the third preset speed threshold;

and if so, determining that the floating car is the floating car with abnormal driving behaviors.

8. The method according to claim 1, wherein the abnormal driving behavior occurrence sections and the easy abnormal driving behavior types in the road network are predetermined by:

training a second road characteristic parameter threshold value corresponding to the specified abnormal driving behavior type by using the historical characteristic data of the road in the training sample set;

when the verification condition is not met, adjusting the threshold value of the second road characteristic parameter, and returning to the step of continuously executing the training;

9. The method of claim 8, further comprising obtaining historical characterization data for the road by:

acquiring historical raw data of roads from road network data, wherein the historical raw data of the roads comprises: at least one of road grade, road type, region where the road is located, traffic light information, toll station information and interest point information of the road;

determining historical dynamic data of a road based on the acquired track data of a plurality of floating cars matched to the road in a preset time period; the historical dynamic data of the road comprises: at least one of a daily average traffic flow of the road, an average speed of the floating car, an average transit time of the floating car, an average parking time of the floating car, and a type of the floating car.

10. The method according to claim 9, wherein the training of the second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type using the historical characteristic data of the road included in the training sample set comprises:

according to the received second road characteristic parameter threshold in training, screening to obtain a road sample meeting the second road characteristic parameter threshold in the training sample set based on historical original data of roads and historical dynamic data of roads of each road sample in the training sample set; the training sample set comprises a plurality of road samples of abnormal driving behavior multiple road sections of specified abnormal driving behavior types;

11. The method of claim 10, the verifying the second road characteristic parameter threshold using historical characteristic data of roads included in a set of verification samples when a training recall condition is met, comprising:

according to the second road characteristic parameter threshold value, screening to obtain a road sample meeting the second road characteristic parameter threshold value in the verification sample set based on historical original data of roads and historical dynamic data of roads in the verification sample set; the verification sample set comprises a plurality of road samples of abnormal driving behavior multi-sending road sections of specified abnormal driving behavior types, and the number of the road samples in the verification sample set is larger than that of the road samples in the training sample set;

12. An abnormal driving behavior road identification method comprises the following steps:

13. A method of road speed determination, comprising:

the floating car identification method according to any one of claims 1 to 11 is used for identifying and acquiring whether the floating car on the running road is a floating car with abnormal running behavior, and determining an abnormal influence factor of the floating car according to the corresponding identification result of the floating car;

14. The method of claim 13, wherein determining the abnormal influence factor of the floating car according to the corresponding recognition result of the floating car comprises:

if the identified floating car has abnormal driving behaviors, determining the abnormal influence factors of the floating car on the road as the weight-reducing influence factors;

if the identified floating car has no abnormal driving behavior, determining the abnormal influence factor of the floating car on the road as the weighting-up influence factor;

if the floating car cannot be identified whether the abnormal driving behavior exists, determining that the abnormal influence factor of the floating car on the road is 1; the impact factor of the falling weight is less than 1 and the impact factor of the rising weight is greater than 1.

15. A real-time road condition determining method comprises the following steps:

determining a traffic speed of a driving road using the road traffic speed determination method of claim 13 or 14;

16. The method of claim 15, wherein determining the real-time traffic status of the driving road based on the traffic speed of the driving road and the type of abnormal driving behavior that is liable to occur on the driving road comprises:

matching the passing speed of the running road with a preset speed threshold value range corresponding to the abnormal running behavior type which is easy to occur on the running road:

if the passing speed of the running road is smaller than the minimum speed threshold value of the preset speed threshold value range, determining that the real-time road condition of the running road is congested;

17. A floating car identification device comprising:

the type acquisition module is used for determining whether the driving road is the abnormal driving behavior multi-occurrence road section or not based on the predetermined abnormal driving behavior multi-occurrence road section in the road network and the easily-occurring abnormal driving behavior type; if yes, acquiring the type of abnormal driving behaviors which are easy to occur in the driving road;

the floating car identification module is used for comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road; and if so, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type.

18. The apparatus of claim 17, further comprising:

the training module is used for training a second road characteristic parameter threshold corresponding to the specified abnormal driving behavior type by using the historical characteristic data of the road included in the training sample set;

the verification module is used for verifying the second road characteristic parameter threshold by using the historical characteristic data of the road in a verification sample set when the training recall condition is met;

19. An abnormal-traveling-behavior road recognition device comprising:

20. A road traffic speed determining apparatus comprising:

the type acquisition module is used for determining whether the driving road is the abnormal driving behavior multi-occurrence road section or not based on the predetermined abnormal driving behavior multi-occurrence road section in the road network and the easily-occurring abnormal driving behavior type; if so, acquiring the type of abnormal driving behaviors which are easy to occur in the driving road;

the floating car identification module is used for comparing the dynamic characteristic data with a first road characteristic parameter threshold value corresponding to the abnormal driving behavior type to determine whether the abnormal driving behavior occurs on the driving road or not; if yes, determining whether the corresponding floating car is the floating car with the abnormal driving behavior or not based on the track data of the floating car on the driving road and the track characteristic parameter threshold value corresponding to the abnormal driving behavior type;

21. A real-time traffic status determination device, comprising:

22. A real-time road condition determination system, comprising: a server and at least one client, wherein:

the server is provided with the real-time road condition determining device of claim 21, and is configured to receive a real-time road condition request sent by the client, and send the determined real-time road condition request to the client;

23. A computer-readable storage medium that, when running, executes at least one of the floating car identification method of any one of claims 1 to 11, the abnormal driving behavior road identification method of claim 12, the road traffic speed determination method of claim 13 or 14, and the real-time road condition determination method of claim 15 or 16.