Method and apparatus for assessing familiarity of a driving route based on gps travel data

Abstract

The embodiment of the application discloses a method for evaluating driving route familiarity based on GPS travel data, comprising the following steps: S1: collecting GPS point data of vehicle historical travel, filtering and travel cutting processing are performed on the GPS point data, and all travels of each vehicle are obtained; S2: for all travels of each vehicle, a DTW algorithm is used to calculate the distance between any two travels; S3: a clustering algorithm is used to cluster the distance, and a clustering result is obtained; S4: according to the clustering result, a Gaussian model is used to evaluate the driving route familiarity of any new travel T. According to the embodiment of the application, the driving route familiarity of each travel can be evaluated, important features can be provided for a model for predicting vehicle accidents and annual compensation amount built based on driving behavior data, and therefore, the driving risk, habit and risk of a driver can be more accurately described.

Description

Technical Field

[0001] This invention relates to the field of vehicle information processing technology, specifically to a method, electronic device, and medium for assessing driving route familiarity based on GPS trip data. Background Technology

[0002] The "Research Report on the Correlation between Driving Behavior Factors in Vehicle Networking Data and Auto Insurance Risk" points out that familiarity with driving routes is highly correlated with the probability of vehicle accidents. Therefore, accurately assessing a driver's familiarity with driving routes is essential.

[0003] Chinese invention patent application CN111460076A discloses a method for determining driving route familiarity, comprising: S201, acquiring a sequence of travel coordinate data corresponding to multiple driving trips of a vehicle; S202, obtaining the trip similarity between the multiple driving trips based on the travel coordinate data sequence; S203, clustering the multiple driving trips based on the trip similarity, and assigning driving trips of the same category to the same driving route; S204, obtaining the driving route familiarity based on the number of driving trips corresponding to each driving route and the total number of driving trips corresponding to the multiple driving trips. The driving route familiarity is calculated using the following method:

[0004]

[0005] In the above formula, f represents the familiarity with the driving route, l represents the number of the driving route to which the driving trip belongs, with a value range of {1, 2, ..., C}, C represents the number of different driving routes, and n l This indicates the number of driving trips corresponding to driving route number 1. Therefore, the driving route familiarity index f represents the proportion of driving trips taken on routes with at least two trips to the total number of driving trips. The value range of this driving route familiarity index f is [0,1]. A higher value indicates greater familiarity with the driving route. This driving route familiarity index f accurately reflects the driver's familiarity with the driving route within a certain period.

[0006] The method for calculating driving route familiarity has the following drawbacks:

[0007] 1. This calculation method can only measure the overall familiarity with all trips within a certain period. Each new trip requires re-running the entire route familiarity determination process before the route familiarity can be calculated. The reason for this is that all trips need to be clustered to determine which category the new trip belongs to, and it may even be possible to change the number of cluster centers before the above formula can be used to calculate familiarity.

[0008] 2. This calculation method cannot calculate the familiarity of a single trip because, as can be seen from the above calculation formula, it calculates the overall familiarity of all historical trips, and therefore cannot distinguish fine-grained trip data.

[0009] 3. This calculation method uses Derivative Dynamic Time Warping (DTW) to calculate the differences between trips, but it does not take into account the actual situation between trip 1 (from point A to point B) and trip 2 (from point B to point A). Having traveled trip 1, one is also somewhat familiar with trip 2. If DTW is used directly for calculation, the difference will be large and does not conform to the actual situation.

[0010] 4. For example, the familiarity with a road taken yesterday is obviously different from that taken a month ago, but this calculation method does not distinguish the importance of all trips in terms of time. Summary of the Invention

[0011] In view of the technical deficiencies mentioned in the background art, the purpose of this invention is to provide a method, electronic device and storage medium for evaluating driving route familiarity based on GPS trip data.

[0012] To achieve the above objectives, in a first aspect, embodiments of the present invention provide a method for assessing driving route familiarity based on GPS trip data, comprising:

[0013] Step S1: Collect GPS point data of vehicle historical trips, filter the GPS point data and segment the trips to obtain all trips of each vehicle;

[0014] Step S2: For all trips of each vehicle, use the DTW algorithm to calculate the distance between any two trips;

[0015] Step S3: Use a clustering algorithm to cluster the distances to obtain the clustering results;

[0016] Step S4: Based on the clustering results, use a Gaussian model to assess the familiarity of driving routes for any new trip T.

[0017] As a specific implementation of this application, the filtering process for GPS point data in step S1 is as follows:

[0018] Filter out GPS point data with repeated latitude and longitude coordinates and a velocity of 0, keeping only one point data;

[0019] Data points with the same timestamp are filtered out, and only one data point is kept.

[0020] Furthermore, in step S1, the trip is segmented using the vehicle status, vehicle speed, and time uploaded by the onboard equipment, specifically as follows:

[0021] Remove all GPS point data when the vehicle is parked, and retain only the trip data when the vehicle is in motion; the vehicle status includes both parked and in motion status.

[0022] Using parking signs, all GPS data collected within a preset time period for a vehicle is divided into multiple trips;

[0023] The trip is segmented using time data from GPS points.

[0024] As a specific implementation of this application, step S2 is as follows:

[0025] The formula D = (d1 + d2 + ... + d) is used. n ) / N calculates the distance between any two travel routes;

[0026] Where d1, d2, d n Let N be the distance between the two closest GPS points in trips A and B. Connect the closest points in trips A and B to form a line segment, where N is the number of line segments.

[0027] Preferably, as a preferred implementation of this application, step S2 specifically includes:

[0028] The formula D = (d1 + d2 + ... + d) is used. n ) / N calculates the first distance between travel A and travel B;

[0029] Reverse the GPS sequence of trip A or trip B using the formula D = (d1 + d2 + ... + d...). n The distance between travel A and travel B is calculated by multiplying the distance by a coefficient greater than 1 to obtain the second distance.

[0030] Take the smaller of the first distance and the second distance as the distance between trip A and trip B;

[0031] Where d1, d2, d n Let N be the distance between the two closest GPS points in trips A and B. Connect the closest points in trips A and B to form a line segment, where N is the number of line segments.

[0032] As a specific implementation of this application, the distance is clustered using the CFDP algorithm or the AP algorithm to obtain the clustering results; the clustering structure includes each cluster center and the category of each trip.

[0033] As a specific implementation of this application, step S4 is as follows:

[0034] Using formula The parameters required to determine the Gaussian model are: mean μ;

[0035] Using formula The parameters required to determine the Gaussian model are: standard deviation σ;

[0036] Substituting the mean μ and standard deviation σ of the parameters into the Gaussian model, the driving route familiarity F is calculated:

[0037]

[0038] Where C is the cluster center, Trip i For the process in this class.

[0039] Furthermore, as a preferred implementation of this application, the method further includes:

[0040] Step S5: Repeat steps S2-S4 to complete the modeling of each vehicle;

[0041] Step S6: Save the clustering results obtained in step S3 and all Gaussian models obtained in step S4.

[0042] Secondly, embodiments of the present invention also provide another electronic device, including a processor, an input device, an output device, and a memory, wherein the processor, input device, output device, and memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to invoke the program instructions to execute the steps of the method described in the first aspect above.

[0043] Thirdly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, implement the steps of the method described in the first aspect above.

[0044] By implementing the embodiments of the present invention, the familiarity with the driving route for each trip can be assessed, which can provide important features for predicting vehicle accidents and annual compensation amounts based on driving behavior data, thereby more accurately depicting the driver's driving risks, habits and risks. Attached Figure Description

[0045] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below.

[0046] Figure 1This is a flowchart of a method for assessing driving route familiarity based on GPS trip data, provided in an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram illustrating the calculation of distances between trips using the DTW algorithm;

[0048] Figure 3 This is a structural diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0051] The key terms involved in the embodiments of this invention are defined as follows:

[0052] GPS (Global Positioning System)

[0053] DTW (Derivative Dynamic Time Warping)

[0054] CFDP (Clustering by fast search and find of density peaks)

[0055] AP (Affinity Propagation) clustering algorithm

[0056] Please refer to Figure 1 The method for assessing driving route familiarity based on GPS trip data provided in this embodiment of the invention includes the following steps:

[0057] Step S1: Collect GPS point data of the vehicle's historical trips, filter the GPS point data, and segment the trips to obtain all the trips of each vehicle.

[0058] Specifically, data filtering mainly includes: (1) Since the parking location is the same as the GPS point, the points with repeated longitude and latitude and speed of 0 are filtered out and only one is kept; (2) Due to the influence of factors such as equipment, satellite signal and point supplementation rules, GPS points are uploaded repeatedly, so GPS points with the same timestamp are filtered out and only one point data is kept.

[0059] Specifically, trip segmentation mainly involves segmenting the trip using vehicle status (including parking and driving status), vehicle speed, and time uploaded by onboard equipment.

[0060] All points in the parked state need to be removed; only trips in the driving state need to be retained. There is a period of time between every two trips where the vehicle is parked at 0 speed. Using the parked state indicator, all GPS points collected by the vehicle within a certain period can be divided into trips, and then further divided based on the time the GPS points were used within each trip. Additionally, if the time between two adjacent GPS points is greater than 10 minutes, the trip is divided again.

[0061] Step S2: For all trips of each vehicle, use the DTW algorithm to calculate the distance between any two trips.

[0062] Specifically, as a particular implementation method, step S2 includes:

[0063] The formula D = (d1 + d2 + ... + d) is used. n ) / N calculates the distance between any two travel routes;

[0064] Where d1, d2, d n Let N be the distance between the two closest GPS points in trips A and B. Connect the closest points in trips A and B to form a line segment, where N is the number of line segments.

[0065] Preferably, as another preferred implementation, the DTW algorithm has been optimized, therefore step S2 includes:

[0066] The formula D = (d1 + d2 + ... + d) is used. n ) / N calculates the first distance between travel A and travel B;

[0067] Reverse the GPS sequence of trip A or trip B using the formula D = (d1 + d2 + ... + d...). n The distance between travel A and travel B is calculated by multiplying the distance by a coefficient greater than 1 to obtain the second distance.

[0068] Take the smaller of the first distance and the second distance as the distance between trip A and trip B;

[0069] Where d1, d2, d n Let N be the distance between the two closest GPS points in trips A and B. Connect the closest points in trips A and B to form a line segment, where N is the number of line segments.

[0070] For example, please refer to Figure 2 In the diagram, trip_old and trip_new represent two trips. The gray lines in the middle connect the points that are closest to each other in the two trips. For example, for three points A, B, and C, the closest point to trip_new from point C is point A, so A and C are connected by a gray line. Similarly, the closest point to trip_new from point B is point A, so A and B are connected. All GPS points between C and B are closest to point A in trip_new. The distance from trip_old to trip_new is the sum of the distances between all the points connected by the gray lines, divided by the number of gray lines.

[0071] As an optimization method, the above approach first calculates the distance between the two routes normally, then reverses the GPS sequence of one of the routes and calculates the distance again. Finally, it multiplies the result by a coefficient greater than 1, and takes the smaller of the two distances as the actual distance between the two routes. This optimization is more realistic; traveling from A to B and then from B to A is relatively familiar, but certainly not as familiar as traveling from A to B again. In other words, the optimized method is more realistic, and the distance between the two routes is more conducive to accurately assessing subsequent driving route familiarity.

[0072] Step S3: Use a clustering algorithm to cluster the distances to obtain the clustering results.

[0073] Specifically, the distances are clustered using either the CFDP algorithm or the AP algorithm. The CFDP algorithm is illustrated below:

[0074] Using the CFDP clustering algorithm requires determining two parameters: density ρ and distance δ. The detailed steps are as follows:

[0075]

[0076] Where ρ i This indicates that when process i is the cluster center, d is used. c The number of passes within a circle of radius d, i.e., density; ij d represents the distance between trips i and j. c For the set distance threshold, d ij Less than d c Then it is assumed that routes i and j belong to the same cluster; χ(x)=1, x<0; χ(x)=0, otherwise;

[0077]

[0078] Where δ i This represents the distance between trip i and trip j, which is the closest trip to trip j among all trips with a density greater than trip i.

[0079] In addition, there will always be a route with the highest density in the data, and the distance δ of this route is directly set to the maximum value.

[0080] δ i =max j (d ij )

[0081] Finally, all vehicle trips are classified according to density ρ and distance δ to obtain each cluster center and the category of each trip.

[0082] Step S4: Based on the clustering results, use a Gaussian model to assess the familiarity of driving routes for any new trip T.

[0083] Based on the clustering results obtained in step 3, a Gaussian model is used to assess the familiarity of a new journey T. The Gaussian model requires determining two parameters: the mean μ and the standard deviation σ. The detailed steps are as follows:

[0084]

[0085] Where C is the cluster center (travel) obtained in step 3. The distance between the new travel and all cluster centers is calculated, and the minimum distance is taken as the mean μ of the Gaussian model.

[0086] The formula for calculating the standard deviation σ is as follows:

[0087]

[0088] Trip i For the processes in this class.

[0089] The above steps yield the mean μ and standard deviation σ of two parameters, which are then substituted into the Gaussian model to calculate the familiarity F.

[0090]

[0091] Step S5: Repeat steps S2-S4 to complete the modeling of each vehicle.

[0092] Step S6: Save the clustering results obtained in step S3 and all Gaussian models obtained in step S4.

[0093] As can be seen from the above description, implementing the embodiments of the present invention can assess the familiarity with the driving route for each trip, and can provide important features for predicting vehicle accidents and annual compensation amounts based on driving behavior data, thereby more accurately depicting the driver's driving risks, habits and risks.

[0094] Optionally, embodiments of the present invention also provide an electronic device. For example... Figure 3 As shown, the aforementioned electronic device may include one or more processors 101, one or more input devices 102, one or more output devices 103, and a memory 104. The processors 101, input devices 102, output devices 103, and memory 104 are interconnected via a bus 105. The memory 104 stores a computer program, which includes program instructions. The processor 101 is configured to invoke the program instructions to execute the method steps described in the foregoing method embodiments.

[0095] It should be understood that, in this embodiment of the invention, the processor 101 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0096] Input device 102 may include a keyboard, etc., and output device 103 may include a display (LCD, etc.), a speaker, etc.

[0097] The memory 104 may include read-only memory and random access memory, and provides instructions and data to the processor 101. A portion of the memory 104 may also include non-volatile random access memory. For example, the memory 104 may also store device type information.

[0098] In specific implementations, the processor 101, input device 102, and output device 103 described in the embodiments of the present invention can execute the implementation methods described in the embodiments of the method for evaluating driving route familiarity based on GPS trip data provided in the embodiments of the present invention, which will not be repeated here.

[0099] It should be noted that for a more detailed description of the workflow of the electronic device, please refer to the aforementioned method embodiment section, which will not be repeated here.

[0100] In addition, corresponding to the aforementioned method embodiments and electronic devices, this embodiment of the invention provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, implement the above-described method for evaluating driving route familiarity based on GPS trip data.

[0101] The computer-readable storage medium can be an internal storage unit of the system described in any of the foregoing embodiments, such as the system's hard disk or memory. The computer-readable storage medium can also be an external storage device of the system, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the system. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0102] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0103] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, or may be electrical, mechanical or other forms of connection.

[0104] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0105] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0106] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0107] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for assessing driving route familiarity based on GPS trip data, characterized in that, include: Step S1: Collect GPS point data of vehicle historical trips, filter the GPS point data and segment the trips to obtain all trips of each vehicle; Step S2: For all trips of each vehicle, use the DTW algorithm to calculate the distance between any two trips; Step S3: Use a clustering algorithm to cluster the distances to obtain the clustering results; Step S4: Based on the clustering results, use a Gaussian model to assess the familiarity of driving routes for any new trip T; Step S2 is as follows: The formula D = (d1 + d2 + ... + d) is used. n ) / N calculates the first distance between travel A and travel B; Reverse the GPS sequence of trip A or trip B using the formula D = (d1 + d2 + ... + d...). n The distance between trips A and B is calculated by multiplying the distance by a coefficient greater than 1 to obtain the second distance. Take the smaller of the first distance and the second distance as the distance between trip A and trip B; Where d1, d2, d n Let N be the distance between the two closest GPS points in trips A and B. Connect the closest points in trips A and B to form a line segment, where N is the number of line segments.

2. The method as described in claim 1, characterized in that, The filtering process for GPS point data in step S1 is as follows: Filter out GPS point data with repeated latitude and longitude coordinates and a velocity of 0, keeping only one point data; Data points with the same timestamp are filtered out, and only one data point is kept.

3. The method as described in claim 1, characterized in that, In step S1, the trip is segmented using the vehicle status, vehicle speed, and time uploaded by the onboard equipment, specifically as follows: Remove all GPS point data when the vehicle is parked, and retain only the trip data when the vehicle is in motion; the vehicle status includes both parked and in motion status. Using parking signs, all GPS data collected within a preset time period for a vehicle is divided into multiple trips; The trip is segmented using time data from GPS points.

4. The method as described in claim 1, characterized in that, In step S3, the distances are clustered using the CFDP algorithm or the AP algorithm to obtain clustering results; the clustering results include each cluster center and the category of each trip.

5. The method as described in claim 4, characterized in that, Step S4 is as follows: Using formula The parameters required to determine the Gaussian model are: mean μ; Using formula The parameters required to determine the Gaussian model are: standard deviation σ; Substituting the mean μ and standard deviation σ of the parameters into the Gaussian model, the driving route familiarity F is calculated: Where C is the cluster center, Trip i For the process in this class.

6. The method according to any one of claims 1-5, characterized in that, Also includes: Step S5: Repeat steps S2-S4 to complete the modeling of each vehicle; Step S6: Save the clustering results obtained in step S3 and all Gaussian models obtained in step S4.

7. An electronic device, characterized in that, The electronic device includes a processor, an input device, an output device, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the steps of the method as described in claim 6.

8. A computer-readable storage medium storing a computer program, the computer program comprising program instructions, characterized in that, When the program instructions are executed by the processor, they implement the steps of the method as described in claim 6.

Images