A method, device and medium for matching people, vehicles and roads based on multi-source information fusion

Abstract

The embodiment of the present invention discloses a method, device and medium for matching people, vehicles and roads based on multi-source information fusion. The method includes: determining the target vehicle through radar and video data of the highway, and identifying the model of each target vehicle according to the ETC card data; selecting different data sources according to the model to determine the original trajectory of each target vehicle; determining the original trajectory of multiple people on the way through mobile phone signaling data within the coverage area of ​​the highway; using the hidden Markov model, matching the original trajectory of each target vehicle and each person on the way to the highway respectively, and obtaining the real trajectory of each target vehicle and each person on the way; for each target vehicle, if there is a person on the way among the multiple people on the way who matches the owner information of the current target vehicle, the person on the way is used as the first person on the way carried by the current target vehicle; determining the second person on the way carried by the target vehicle according to the trajectory similarity and speed similarity.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of intelligent transportation technology, and in particular to a method, device and medium for matching a person, a vehicle and a road based on multi-source information fusion. Background Art

[0002] With the rapid development of expressways and the increase in vehicles, the use and operation of expressways have encountered more and more problems. For example, it is impossible to obtain the specific information of the people in the car in time, and rescue cannot be carried out in time if an accident occurs. For example, when the vehicle is not equipped with the GPS Beidou positioning system or the system fails, it is difficult to track the vehicle in time.

[0003] In the existing technology, human-vehicle tracking technology can be achieved through human-vehicle matching, but due to errors in the positions of people and vehicles, it is impossible to accurately locate the road network, and it is difficult to accurately infer the position of one party from the position of the other party. Summary of the invention

[0004] The embodiments of the present invention provide a method, device and medium for matching people, vehicles and roads based on multi-source information fusion, which realizes accurate matching of people, vehicles and roads on highways based on multi-source big data such as road data, mobile phone signaling data, ETC data, checkpoint data and GPS Beidou data.

[0005] In a first aspect, an embodiment of the present invention provides a method for matching a person, a vehicle and a road based on multi-source information fusion, including:

[0006] Determine at least one target vehicle through radar and video data of the highway, and identify the model of each target vehicle based on ETC card data; select different data sources based on the model to determine the original trajectory of each target vehicle;

[0007] Determine the original trajectories of multiple people on the road through mobile phone signaling data within the coverage area of ​​the highway;

[0008] Using the hidden Markov model, the original trajectories of each target vehicle and each person on the road are matched to the highway respectively, and the real trajectories of each target vehicle and each person on the road are obtained;

[0009] For each target vehicle, perform the following operations:

[0010] S1. If there is a person on the way among the multiple persons on the way who matches the owner information of the current target vehicle, the person on the way is regarded as the first person on the way carried by the current target vehicle;

[0011] S2. Calculate the trajectory similarity between the current target vehicle and the other multiple on-the-go persons based on the Fréchet distance and the sum of squared deviations between the real trajectories; select at least one on-the-go person with the highest trajectory similarity as at least one candidate person carried by the current target vehicle;

[0012] S3. Calculate the speed of the current target vehicle and each candidate person in the same time period, select the candidate person with the highest speed similarity as the second person on the way carried by the target vehicle, and the first person on the way and the second person on the way together constitute the set of people carried by the target vehicle.

[0013] In a second aspect, an embodiment of the present invention provides an electronic device, the electronic device comprising:

[0014] one or more processors;

[0015] a memory for storing one or more programs,

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the human-vehicle-road matching method based on multi-source information fusion described in any embodiment.

[0017] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the human-vehicle-road matching method based on multi-source information fusion as described in any embodiment.

[0018] The embodiments of the present invention can achieve the following beneficial effects:

[0019] 1. When considering road section matching, the embodiment of the present invention integrates multi-source data and uses a hidden Markov model to perform human-road matching and vehicle-road matching, integrates mobile phone signaling data and road data to perform human-road matching to obtain the movement trajectory of people on the highway, integrates GPS, Beidou data and road data to perform vehicle-road matching to obtain the trajectory of passenger and freight vehicles, and integrates checkpoint data and road data to perform vehicle-road matching to obtain the trajectory of sedans, so that trajectory recognition and human-vehicle-road matching have strong robustness.

[0020] 2. The embodiment of the present invention calculates the spatiotemporal similarity of the trajectories of people and vehicles on the road through a combination algorithm of Fréchet distance and sum of squared deviation distance, and combines the speed information of people and vehicles to match people with their vehicles, thereby improving the success rate of matching people and vehicles.

[0021] 3. The embodiments of the present invention can alleviate the problems in the prior art of being unable to timely and effectively obtain the situation of people in the car in a traffic accident, being unable to accurately send the identified traffic status to the mobile phones of people on the highway, and being unable to accurately track people and vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a flow chart of a method for matching a person, a vehicle and a road based on multi-source information fusion provided by an embodiment of the present invention;

[0024] Figure 2 is a schematic diagram of vehicle-road matching of a truck provided by an embodiment of the present invention;

[0025] Figure 3 is a schematic diagram of vehicle-road matching of a car provided by an embodiment of the present invention;

[0026] Figure 4 This is a schematic diagram of person-road matching for people on the road provided by an embodiment of the present invention;

[0027] Figure 5 This is another schematic diagram of person-road matching for people on the road provided by an embodiment of the present invention;

[0028] Figure 6 This is another schematic diagram of person-road matching for people on the road provided by an embodiment of the present invention;

[0029] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0031] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0032] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0033] Figure 1 1 is a flow chart of a method for matching a person, vehicle and road based on multi-source information fusion provided by an embodiment of the present invention. The method is applicable to matching a person, vehicle and road on a highway and is executed by an electronic device. Figure 1 As shown, the method specifically includes:

[0034] S110, determining at least one target vehicle through radar and video data of the highway, and identifying the model of each target vehicle according to the ETC card data; selecting different data sources according to the model to determine the original trajectory of each target vehicle.

[0035] This embodiment determines the original trajectory of each vehicle passing through the highway through multi-source data, and the multi-source data includes radar data, video data, vehicle-mounted GPS data, Beidou vehicle-mounted satellite data, etc. This embodiment uses different data sources to determine the original trajectories of different vehicles in different stages and vehicle types according to the application scenarios and characteristics of various data sources, thereby improving the robustness and accuracy of the original trajectories.

[0036] In a specific implementation, first, the license plate and appearance features of vehicles on the highway are identified through radar data on the highway and video data from ETC gantries and cameras. The identification results are uploaded to the vehicle information database of the vehicle management department, and the vehicles that match the matching can be used as target vehicles for human-vehicle-road matching.

[0037] Then, the model of each target vehicle is identified according to the ETC card data. Optionally, the vehicle types mainly include two-passenger and one-hazardous vehicle, commercial truck and sedan, etc. If the target vehicle is a two-passenger and one-hazardous vehicle or commercial truck, the vehicle is equipped with GPS and Beidou vehicle-mounted satellite positioning, which has high accuracy in detecting the vehicle's spatiotemporal information. The vehicle position can be detected in real time through the vehicle-mounted GPS or Beidou vehicle-mounted satellite; if the target vehicle is a car, the vehicle position is detected through the checkpoint data. The detected vehicle position includes the vehicle's latitude and longitude and the corresponding time information, so that the original vehicle trajectory data set of the highway can be established.

[0038] Next, the collected vehicle trajectory data set is processed for abnormal data, mainly including the removal of missing data and the cleaning of duplicate data. Optionally, data with missing key field information such as longitude and latitude can be directly deleted; for data with multiple fields that are the same, only one field can be retained.

[0039] Finally, the trajectory data points with the same license plate number are extracted from the processed data, and each license plate number corresponds to a target vehicle. The original trajectory points of each target vehicle X are arranged in chronological order to form the original trajectory point sequence of the target vehicle X (P1…P i …P n ).

[0040] S120. Determine the trajectories of multiple persons on the road through mobile phone signaling data within the coverage area of ​​the expressway.

[0041] In this embodiment, while determining the vehicle trajectory, the trajectory of people within the expressway is also determined through mobile phone signaling data. In a specific implementation, first, the longitude and latitude of all mobile phone users within the coverage of base stations along the expressway and the time information corresponding to the location are obtained to establish the original mobile phone user trajectory dataset of the expressway.

[0042] Then, the collected mobile user trajectory data set is processed for abnormal data, mainly including the elimination of missing data and the cleaning of duplicate data. Optionally, data with missing key field information such as longitude and latitude can be deleted directly; for some signaling events, multiple data with the same fields except time will be generated in a short time (several seconds). This type of data is duplicate data generated by the system during transmission, storage, etc. It is not effective in improving the precision and accuracy of travel research results, and will increase the workload and complexity of data processing. Therefore, it can be filtered and one of the data can be retained.

[0043] Finally, the trajectory data points with the same user code are extracted and arranged in chronological order to form the original trajectory point sequence (S1…S j …S m ).

[0044] S130. Using a hidden Markov model, the original trajectories of the target vehicles and the on-the-way personnel are matched to the highways to obtain the real trajectories of the target vehicles and the on-the-way personnel.

[0045] In actual applications, there are errors in the collected vehicle and personnel positions, and the original trajectories of each target vehicle and the original trajectories of the personnel may deviate from the highway. In order to improve the accuracy of the trajectory of the person and the vehicle, the road network data can be used to modify the original trajectory to match the highway where the person and the vehicle pass.

[0046] Optionally, for each target vehicle, the original trajectory sequence of the vehicle is matched to the corresponding highway section through a hidden Markov model to obtain the real trajectory point sequence of the vehicle (V1…V i …V n ), connect the matching trajectory points, and you can get the vehicle's driving trajectory on the highway, which is called the true trajectory. For example, the matching process of a truck is as follows Figure 2 As shown in Figure 2, the matching process of the car is as follows: Figure 3 As shown in the figure, the horizontal axis represents longitude, the vertical axis represents latitude, the points connected by dotted lines represent the original trajectory point sequence, and the points connected by solid lines represent the real trajectory point sequence. Each trajectory point can be represented as follows:

[0047] P i = {llon i ,llat i ,tt i},V i = {lon i ,lat i ,t i}, where P i is the i-th original trajectory point of the target vehicle arranged in time order, llon i is the longitude of the i-th original trajectory point of the target vehicle, llat i is the latitude of the i-th original trajectory point of the target vehicle, tt i is the time corresponding to the original position of the target vehicle, which can be accurate to seconds; V i is the i-th real trajectory point of the target vehicle arranged in time order, lon i is the longitude of the i-th real trajectory point of the target vehicle, lat i is the latitude of the i-th real trajectory point of the target vehicle, t i is the time corresponding to the original position of the target vehicle, which can be accurate to seconds; n is the number of driving trajectory points of target vehicle k, X=1…k, k is the number of target vehicles in the vehicle trajectory dataset.

[0048] Optionally, for each person on the way, the original trajectory sequence of the person on the way is matched to the corresponding highway section through the hidden Markov model to obtain the real trajectory point sequence of the person on the way (U1…U j …U m ), connect the matching trajectory points, and thus obtain the movement trajectory of the person on the road on the highway, also known as the real trajectory. For example, the trajectory matching process of several people on the road is as follows Figure 4 , Figure 5 , Figure 6 As shown in the figure, the horizontal axis represents longitude, the vertical axis represents latitude, the points connected by dotted lines represent the original trajectory point sequence, and the points connected by solid lines represent the real trajectory point sequence. Each trajectory point can be represented as follows:

[0049] S j = {llon j ,llat j ,tt j},U j = {lon j ,lat j ,t j}, where S j is the jth original trajectory point of the person on the way arranged in chronological order, j is the longitude of the jth original trajectory point of the person on the way, llat j is the latitude of the jth original trajectory point of the person on the way, tt j The time corresponding to the original position of the person in transit, which can be accurate to seconds; U j is the jth real trajectory point of the person on the way arranged in chronological order, lon j is the longitude of the jth real track point of the person on the way, lat j is the latitude of the jth real trajectory point of the person on the way, t j is the time corresponding to the real position of the person in transit, which can be accurate to seconds; m is the number of Y motion trajectory points of the person in transit, Y=1…g, g is the number of people in transit in the mobile phone user trajectory data set.

[0050] S140: For each target vehicle, perform the following operations respectively:

[0051] S1. If there is a person on the way among the multiple persons on the way who matches the owner information of the current target vehicle, the person on the way is taken as the first person on the way carried by the target vehicle.

[0052] For each target vehicle, this embodiment first determines whether the owner is a person on the way. Optionally, first, after uploading the vehicle identification information to the vehicle management department in S110, the owner information of the target vehicle can be searched from the vehicle information database of the vehicle management department, including the owner's mobile phone number. Then, the mobile phone number is searched from the mobile phone signaling data obtained in S120. If the existing mobile phone number is the same as the mobile phone number in the current target vehicle owner information, the person corresponding to the mobile phone number is identified as the owner of the target vehicle and as a person on the way carried by the target vehicle. If it does not exist, the match fails, and the vehicle owner is not a person on the way.

[0053] S2. Calculate the trajectory similarity between the current target vehicle and the other multiple on-the-go persons according to the Fréchet distance and the sum of squared deviations between the real trajectories; and select at least one on-the-go person with the highest trajectory similarity as at least one candidate person carried by the current target vehicle.

[0054] This embodiment uses a combination of the Fréchet distance and the sum of squared deviation distance to measure the spatiotemporal similarity between the trajectories of the vehicle and the person on the road, as a basis for person-vehicle matching.

[0055] In a specific implementation, before calculating the trajectory similarity, the approximate similarity of the obtained vehicle and person trajectories can be observed, and the persons with similar trajectories to vehicle X can be selected to form a candidate set A of persons carried by vehicle X, thereby reducing the complexity of the calculation. Optionally, target vehicles and persons on the way with approximately the same time range and spatial range can be screened to form a candidate set A of each vehicle, and trajectories with too large differences can be excluded.

[0056] Then, the mean of the Fréchet distance and the mean of the squared deviation distance are calculated to measure the spatiotemporal similarity between the real trajectories of people and vehicles. Among them, the Fréchet distance is suitable for describing the spatiotemporal relationship between trajectories. It takes into account the distance between adjacent points on the trajectory, and while considering the time order, it also pays attention to the similarity of spatial positions. It is very effective in processing the shape of the trajectory and the overall shape change of the path; while the squared deviation distance measures the difference in position between the two trajectories at each moment. By calculating the sum of the squares of the Euclidean distances between the corresponding points on the trajectory, it can very sensitively capture the differences in details. By combining these two distance metrics, the overall shape of the trajectory and the specific position at each moment can be comprehensively considered, which helps to overcome the limitations that may exist in a single distance metric, making the measurement of spatiotemporal similarity more comprehensive and accurate, thereby better capturing the spatiotemporal relationship between the trajectories of vehicles and people on the road. Specifically, the process can include the following steps:

[0057] Step 1: Determine a first track with fewer track points and a second track with more track points in the real track of the current target vehicle and any other person on the road. That is, the real track of the current target vehicle and the real track of any other person on the road, the one with fewer track points is used as the first track, and the other is used as the second track, and the points in the first track are called first track points, and the points in the second track are called second track points.

[0058] Step 2: Find the second trajectory points in the second trajectory that are closest in time to the first trajectory points in the first trajectory and pair them to obtain multiple point pairs; then arrange the point pairs in chronological order to form a point pair sequence. Optionally, each point pair is recorded as (V i ,U j ), where i=j, and the point pair sequence is recorded as L(XY):

[0059] L(XY)={(V1,U1),(V2,U2),…,(V N ,U N )}

[0060] Step 3: Calculate the time-space distance of each point pair in the point pair sequence respectively; accumulate the time-space distances and average them to obtain the sum of squared deviations between the current target vehicle and any other real trajectory of the person on the way. Optionally, in this embodiment, any two points V in the real trajectory point sequence of the vehicle and the real trajectory point sequence of the person on the way are i (lon i ,lat i ,t i ) and U j (lon j ,lat j ,t j ) is defined as the Euclidean distance between two points in the trajectory point sequence arranged in time order:

[0061]

[0062] Based on the above definition, assuming that the real trajectory of the vehicle is the first trajectory, the temporal and spatial distances of each point pair in the point pair sequence are calculated and accumulated to obtain the average, which is equivalent to accumulating the distance between each point in the real trajectory point sequence of the vehicle and the closest point in the real trajectory point sequence of the person on the way, and calculating the average value. The calculation method is as follows:

[0063]

[0064] Wherein, N represents the number of first trajectory points.

[0065] Step 3: For each first trajectory point in the point pair sequence, calculate the maximum spatiotemporal distance between the current first trajectory point and each second trajectory point in the point pair sequence, and use the minimum value of the maximum spatiotemporal distances corresponding to each first trajectory point as the Fréchet distance between the current target vehicle and the real trajectory of any other person on the road. Assuming that the real trajectory of the vehicle is the first trajectory,

[0066] For each vehicle trajectory point in the point pair sequence, the temporal and spatial distance M (V i ,U j ), and take all M(V i ,U j ) as the maximum space-time distance ‖L‖ corresponding to the vehicle trajectory point i :

[0067]

[0068] That is, ‖L‖ represents the maximum value of the temporal and spatial distance between possible sequence pairs when i is determined, and the Fréchet distance between trajectories is:

[0069] δ F (X,Y)=min(‖L‖1,…‖L‖ i ,…‖L‖ N )

[0070] Similarly, when the actual trajectory of the person on the road is the first trajectory, it is only necessary to swap the actual trajectory of the target vehicle with the actual trajectory of the person on the road and then calculate. The specific method is similar to the above process and will not be repeated.

[0071] Step 4: Weighted average the deviation square sum distance and Fréchet distance and normalize them to obtain the trajectory similarity between the current target vehicle and any other person on the way. In order to reduce the error, this embodiment uses the mean value D of the Fréchet distance and the deviation square sum distance to measure the spatiotemporal similarity between the real trajectories of the vehicle and the person on the way. Optionally, the calculation method is as follows:

[0072]

[0073] Furthermore, in order to obtain normalized trajectory similarity, the distance-based measurement results can also be normalized so that the results are within the interval [0,1] for easy comparison and analysis. Optionally, the normalization formula is:

[0074]

[0075] Among them, S represents the normalized trajectory similarity.

[0076] Finally, at least one person on the way with the highest trajectory similarity is selected as at least one candidate person carried by the current target vehicle. That is, from the candidate set A of persons carried by vehicle X, the persons on the way with the highest trajectory similarity to vehicle X are further selected to form the candidate set B of persons carried by vehicle X.

[0077] S3. Calculate the speed of the current target vehicle and each candidate person in the same time period; select the candidate person with the highest speed similarity as the second person on the way carried by the target vehicle, and the first person on the way and the second person on the way together constitute the set of people carried by the target vehicle.

[0078] This embodiment combines the trajectory information and speed information of vehicles and people to match people on the highway with their vehicles. In a specific implementation, first, the speeds of the target vehicle and the people on the highway are calculated. Optionally, for the target vehicle and any candidate person in the candidate set B, the speed calculation process may include the following steps:

[0079] Step 1: Divide the real trajectory of the target vehicle and the real trajectory of the candidate into at least one time period; for each same time period, the movement distance d of the target vehicle or the candidate can be calculated according to the following formula: i ;

[0080]

[0081] Among them, lat i and lon i Respectively represent the starting longitude and latitude of the current target vehicle or any candidate in the same period, lat i+1 and lon i+1 They respectively represent the latitude and longitude of the end point of the current target vehicle or any candidate person in the said time period, and R represents the radius of the earth.

[0082] Step 2: Based on the above movement distance, the interval speed v between the candidate and the target vehicle X at two points on the road section in the same period can be calculated. i , the method is as follows:

[0083] v i =d i / t i

[0084] Among them, t i Represents time intervals of the same duration.

[0085] Step 3: After obtaining the interval speed in each time period, add up the interval speeds of each trajectory in each time period and calculate the average to obtain the average section speed of the entire trajectory.

[0086] Step 4: After performing the above operations on each candidate in the candidate set B, the average road speed of each candidate can be obtained, and the average road speed of the target vehicle X is matched respectively, and at least one candidate with the highest similarity to the average road speed of the target vehicle is screened out, and it is determined that these people are on the same vehicle, and the vehicle is X. For the convenience of distinction and description, in this embodiment, the on-the-way person carried by the target vehicle determined in S1 is called the first on-the-way person (vehicle owner), and the on-the-way person carried by the target vehicle determined in S3 is called the second on-the-way person (non-vehicle owner), and the first on-the-way person and the second on-the-way person together constitute the on-the-way person set carried by the target vehicle.

[0087] In summary, this embodiment provides a human-vehicle-road matching method based on multi-source information fusion, which can achieve the following beneficial effects:

[0088] 1. When considering road section matching, this embodiment integrates multi-source data and uses hidden Markov model to perform human-road matching and vehicle-road matching. It integrates mobile phone signaling data and road data to perform human-road matching to obtain the movement trajectory of people on the highway. It integrates GPS, Beidou data and road data to perform vehicle-road matching to obtain the trajectory of passenger and freight vehicles. It integrates checkpoint data and road data to perform vehicle-road matching to obtain the trajectory of sedan vehicles, so that trajectory recognition and human-vehicle-road matching have strong robustness.

[0089] 2. This embodiment uses a combination algorithm of Fréchet distance and sum of squared deviation distance to calculate the spatiotemporal similarity of the trajectories of people and vehicles on the road, and combines the speed information of people and vehicles to match people with their vehicles, thereby improving the success rate of matching people and vehicles.

[0090] 3. This embodiment can alleviate the problems in the prior art of being unable to timely and effectively obtain the situation of people in the car in a traffic accident, being unable to accurately send the identified traffic status to the mobile phones of people on the highway, and being unable to accurately track people and vehicles.

[0091] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention is shown in FIG. Figure 7 As shown, the device includes a processor 60, a memory 61, an input device 62 and an output device 63; the number of processors 60 in the device can be one or more. Figure 7 A processor 60 is taken as an example; the processor 60, the memory 61, the input device 62 and the output device 63 in the device can be connected by a bus or other means. Figure 7 The example of connecting through bus is taken in the following.

[0092] The memory 61 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the human-vehicle-road matching method based on multi-source information fusion in the embodiment of the present invention. The processor 60 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 61, that is, realizing the above-mentioned human-vehicle-road matching method based on multi-source information fusion.

[0093] The memory 61 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 61 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 61 may further include a memory remotely arranged relative to the processor 60, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0094] The input device 62 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device 63 may include a display device such as a display screen.

[0095] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the human-vehicle-road matching method based on multi-source information fusion of any embodiment is implemented.

[0096] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0097] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0098] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0099] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A human-vehicle-road matching method based on multi-source information fusion, characterized in that: include: Determine at least one target vehicle through radar and video data of the highway, and identify the model of each target vehicle based on ETC card data; select different data sources based on the model to determine the original trajectory of each target vehicle; Determine the original trajectories of multiple people on the road through mobile phone signaling data within the coverage area of ​​the highway; Using the hidden Markov model, the original trajectories of each target vehicle and each person on the road are matched to the highway respectively, and the real trajectories of each target vehicle and each person on the road are obtained; For each target vehicle, perform the following operations: S1. If there is a person on the way among the multiple persons on the way who matches the owner information of the current target vehicle, the person on the way is regarded as the first person on the way carried by the current target vehicle; S2, calculating the trajectory similarity between the current target vehicle and the other multiple on-the-way persons based on the Fréchet distance and the sum of squared deviations between the real trajectories; Selecting at least one on-the-go person with a top trajectory similarity as at least one candidate person carried by the current target vehicle; S3. Calculate the speed of the current target vehicle and each candidate person in the same time period, select the candidate person with the highest speed similarity as the second person on the way carried by the target vehicle, and the first person on the way and the second person on the way together constitute the set of people carried by the target vehicle.

2. The method according to claim 1, characterized in that The method of determining at least one target vehicle through radar and video data of the highway includes: Identify the license plates and appearance features of vehicles on highways through radar data on highways and video data from ETC gantries and cameras; The identification results are uploaded to the vehicle information database of the vehicle management department, and at least one target vehicle is determined through comparison.

3. The method according to claim 1, characterized in that: The target vehicles include two passenger and one hazardous vehicle, commercial trucks and sedans; The method of selecting different data sources according to the vehicle type to determine the original trajectory of each target vehicle includes: If a target vehicle is a two-passenger and one-hazardous vehicle or an operational truck, the vehicle position at different times is determined by the vehicle-mounted GPS or Beidou vehicle-mounted satellite; If a target vehicle is a car, the vehicle position at different times is detected through the bayonet data; The vehicle positions of the target vehicle at different times are arranged as an original trajectory of the target vehicle.

4. The method according to claim 1, characterized in that The original trajectories of multiple persons on the road are determined by using the mobile phone signaling data within the coverage area of ​​the expressway, including: Mobile phone signaling data with the same user code is identified as mobile phone signaling data of the same person on the way; The positions of the person on the way at different times are extracted from the mobile phone signaling data and arranged as the original trajectory of the person on the way.

5. The method according to claim 1, characterized in that If there is a person on the way among the multiple persons on the way that matches the owner information of the current target vehicle, taking the person on the way as the first person on the way carried by the current target vehicle includes: If there is a person on the way among the multiple persons on the way whose mobile phone number is the same as the mobile phone number in the owner information of the current target vehicle, the person on the way will be identified as the owner of the current target vehicle and as the first person on the way carried by the current target vehicle.

6. The method according to claim 1, characterized in that The method of calculating the trajectory similarity between the current target vehicle and the other multiple persons on the way based on the Fréchet distance and the sum of squared deviations between the real trajectories includes: Determine a first trajectory with fewer trajectory points and a second trajectory with more trajectory points in the real trajectory of the current target vehicle and any other on-the-way person; find the second trajectory points closest in time to each first trajectory point in the first trajectory from the second trajectory to pair them up, and obtain a point pair sequence arranged in chronological order; Calculate the time-space distance of each point pair in the point pair sequence respectively; accumulate and average the time-space distances to obtain the sum of squared deviations between the real trajectories of the current target vehicle and any other on-the-way person; For each first trajectory point in the point pair sequence, respectively calculate the maximum spatiotemporal distance between the current first trajectory point and each second trajectory point in the point pair sequence, and use the minimum value of the maximum spatiotemporal distances corresponding to each first trajectory point as the Fréchet distance between the current target vehicle and the real trajectory of any other on-the-way person; The squared deviation distance and the Fréchet distance are weighted and averaged and normalized to obtain the trajectory similarity between the current target vehicle and any other on-the-way person.

7. The method according to claim 1, characterized in that The weighted averaging and normalizing the sum of squared deviations and the Fréchet distance comprises: The weighted average distance D is normalized according to the following formula: Among them, S represents the normalized trajectory similarity.

8. The method according to claim 1, characterized in that The calculating the speed of the current target vehicle and each candidate person in the same time period includes: The movement distance d of the current target vehicle or any candidate in the same period is calculated according to the following formula i ; Among them, lat i and lon i Respectively represent the starting longitude and latitude of the current target vehicle or any candidate in the same period, lat i+1 and lon i+1 respectively represent the latitude and longitude of the destination of the current target vehicle or any candidate in the same time period, and R represents the radius of the earth; The interval speed v of the current target vehicle or any candidate in the same time period is calculated according to the following formula i : Among them, t i Indicates the time interval of the same period.

9. An electronic device, characterized in that: include: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the human-vehicle-road matching method based on multi-source information fusion as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the human-vehicle-road matching method based on multi-source information fusion as described in any one of claims 1-8 is implemented.

Images