A long-distance truck common route identification method based on GPS data

Abstract

The application discloses a truck frequently traveled route identification method based on GPS data, and comprises the following steps: acquiring vehicle networking data through satellite positioning data provided by vehicle networking collection equipment; processing the data in a travel mode, combining discrete GPS points into multiple travel sections through an algorithm; testing the travel data and filtering low-quality travel; counting the prefecture-level cities of the starting points and ending points of the filtered travel; calculating the distance between any two cities where the truck has stayed; constructing an inter-city network flow for the travel trajectory of each truck; and performing a frequently traveled route sorting algorithm. According to the travel trajectory of each truck, a network flow is established, and the frequently traveled route of the truck is predicted, and meanwhile, the problem of inaccurate identification of the frequently traveled route caused by missing data collection and too long resting time of the truck driver during the journey can be solved, and the problem of inaccurate identification of the frequently traveled route caused by the selection of different passing cities by the truck when going back and forth between two prefecture-level cities due to loading and unloading requirements can also be solved.

Description

Technical Field

[0001] This invention relates to a method for identifying frequently used routes of long-haul trucks, specifically a route sorting method based on GPS data and network flow, belonging to the field of positioning systems and data processing technology. Background Technology

[0002] GPS (Global Positioning System) provides accurate positioning for over 98% of the Earth's surface. In recent years, the continuous development of my country's vehicle-to-everything (V2X) industry and the gradual maturation of V2X data collection equipment have enabled the large-scale and real-time acquisition of truck driving trajectories.

[0003] However, during the truck's operation, data collection may be incomplete or drivers may take extended breaks, load, or unload cargo. These data gaps or driver actions during the journey make it difficult to determine the actual start and end points of each trip based solely on the driver's engine start-up and shutdown actions, thus hindering the identification of the truck's frequently used routes. Summary of the Invention

[0004] The purpose of this invention is to provide a method for identifying frequently traveled routes of trucks based on GPS data and network flow sorting algorithms in order to solve the problem.

[0005] This invention achieves the above objective through the following technical solution: a method for identifying frequently traveled routes of trucks based on GPS data, comprising the following steps:

[0006] Step 1: Acquire vehicle network data by using satellite positioning data provided by vehicle network acquisition equipment;

[0007] Step 2: Data route conversion, which uses algorithms to merge discrete GPS points into multiple route segments;

[0008] Step 3: Examine the trip data and filter out low-quality trips;

[0009] Step 4: Statistically filter the cities where the origin and destination of the trip are located;

[0010] Step 5: Calculate the distance between any two cities you have visited;

[0011] Step 6: Construct an intercity network flow for the travel trajectory of each vehicle;

[0012] Step 7: Sort algorithm for frequently used routes.

[0013] As a further technical solution of the present invention: in step one: the vehicle network data collection equipment includes a pre-installed device T-Box, a rear-end data collection device OBD, a rearview mirror and a driving recorder; the satellite positioning data includes satellite positioning longitude, satellite positioning latitude, satellite positioning time, system equipment time and satellite accuracy.

[0014] As a further technical solution of the present invention: the data runout processing in step two includes:

[0015] A. Given all the GPS data points collected from a truck, sort them in ascending order by satellite positioning time;

[0016] B. Calculate the system equipment time interval between each collection point and the previous point. If the interval is greater than 10 minutes, then the point is regarded as the starting point of the next segment of the journey.

[0017] C. Mark the satellite positioning time of the first point of each trip segment as the start time of the trip, and mark the satellite positioning time of the last point of each trip segment as the end time of the trip;

[0018] D. Mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the first point of each segment as the starting city of the segment, and mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the last point of each segment as the ending city of the segment;

[0019] E. Calculate the spherical distance between every two adjacent points using the semi-versus formula. The sum of these values ​​is recorded as the distance traveled on the journey, where R is the Earth's radius.

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] F. Subtract the start time of the trip from the end time of the trip to record the travel time.

[0026] As a further technical solution of the present invention: In step three, low-quality strokes are filtered according to the following rules:

[0027] A. Filter out trips with fewer than 10 satellite points;

[0028] B. Filter out trips with a mileage of less than 300 meters;

[0029] C. Filter trips with a travel time of less than 10 minutes.

[0030] As a further technical solution of the present invention: In step four:

[0031] 1) Given a truck, the route used is the route filtered in step three;

[0032] 2) After statistical filtering, calculate the prefecture-level cities where the starting and ending points of all the truck's trips are located, and record the average longitude of the points belonging to the same prefecture-level city as the longitude of the main stop of the truck in that prefecture-level city. Calculate the average latitude of the points belonging to the same prefecture-level city as the latitude of the main stop of the truck in that prefecture-level city.

[0033] As a further technical solution of the present invention: In step five, given a long-distance truck, the distance between any two different prefecture-level cities where the truck stops is calculated. The specific steps for calculating the distance between two different prefecture-level cities are as follows:

[0034] A. Select two different prefecture-level cities from the cities of stay counted in step four;

[0035] B. The average value of the satellite positioning latitude and longitude calculated in step four represents the latitude and longitude coordinates of the prefecture-level city. That is, the calculated average value of the satellite positioning represents the center of the truck's stop in the prefecture-level city.

[0036] C. The semi-sine formula is used to calculate the spherical distance between two cities, thus calculating the distance between the two locations.

[0037] As a further technical solution of the present invention: In step six, given the travel data of a long-haul truck, the network flow of the truck is constructed, all possible routes are searched, and the flow of each route is calculated. The specific steps are as follows:

[0038] 1) Record each prefecture-level city counted in step four as a node in the network flow, arc This indicates that the truck's journey included a starting point in a prefecture-level city. The destination is a prefecture-level city. The itinerary, for The capacity indicates that the starting point is a prefecture-level city. The destination is a prefecture-level city. Number of trips;

[0039] 2) Path search: Given any two distinct nodes in a network flow and Let's assume Starting from a prefecture-level city, For the destination prefecture-level city, search for nodes in the network flow. To the node All paths, where each path does not contain duplicate nodes;

[0040] 3) The number of filtered path nodes exceeds the threshold ( The excessively long path from the prefecture-level city is obtained. to prefecture-level cities All path sets ;

[0041] 4) Calculation path Maximum capacity Its value is the minimum radian capacity of all radians in the path.

[0042] .

[0043] 5) Calculate the results for any two different prefecture-level cities. and All capacities between

[0044]

[0045]

[0046] .

[0047] As a further technical solution of the present invention: In step seven, all paths are sorted, and the specific steps are as follows:

[0048] 1) Based on the path length calculated in step five, sort all possible routes in descending order of distance from the start and end points, and select the route with the longest distance. The route, the result is ;

[0049] 2) Based on the path capacity calculated in step six, In The routes are sorted in descending order, and the result is: ;

[0050] 3) According to The result determines the threshold for path capacity. ,choose Routes with a capacity greater than this threshold are reordered in descending order of distance between the start and end points, resulting in: ;

[0051] 4) Select The front of the middle The route serves as the long-haul truck's... A frequently used route.

[0052] The beneficial effects of this invention are:

[0053] 1) Using the average latitude and longitude of each long-haul truck at each stop on satellite positioning to represent the latitude and longitude coordinates of the prefecture-level city corresponding to each truck can more accurately calculate the mileage of each truck between two prefecture-level cities.

[0054] 2) This invention constructs a corresponding network flow based on the driving trajectory of long-haul truck drivers and searches all possible routes. Compared with the traditional method of segmenting the trip solely based on stop time, it can solve the following problems: a) inaccurate identification of frequently used routes due to missing data collection and excessively long rest stops by truck drivers (e.g., exceeding 8 hours); b) inaccurate identification of frequently used routes due to trucks choosing different transit cities when traveling between two prefecture-level cities for loading and unloading needs. Furthermore, this invention integrates distance and traffic flow to sort all possible routes, enhancing the accuracy of the route identification algorithm. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0056] Figure 2 This is a schematic diagram of the network flow generated after a given trip of a truck according to the present invention;

[0057] Figure 3 This is a pseudocode diagram of the path search algorithm of the present invention; Detailed Implementation

[0058] 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 embodiments of the present invention, and not all embodiments. 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.

[0059] Example 1

[0060] A method for identifying frequently used truck routes based on GPS data includes the following steps:

[0061] First: Acquire vehicle network data by using satellite positioning data provided by vehicle network data collection equipment.

[0062] In step one, the vehicle network data collection equipment includes a pre-installed T-Box, a rear-end OBD device, a rearview mirror, and a dashcam; the satellite positioning data includes satellite positioning longitude, satellite positioning latitude, satellite positioning time, system equipment time, and satellite accuracy. Common data point collection frequencies are 1 second, 15 seconds, or 30 seconds.

[0063] Second: Data route processing, which uses algorithms to merge discrete GPS points into multiple route segments.

[0064] In step two, the data runout processing includes:

[0065] A. Given all the GPS data points collected for a truck, sort them in ascending order by satellite positioning time;

[0066] B. Calculate the system equipment time interval between each collection point and the previous point. If the interval is greater than 10 minutes, then the point is regarded as the starting point of the next segment of the journey.

[0067] C. Mark the satellite positioning time of the first point of each trip as the start time of the trip, and mark the satellite positioning time of the last point of each trip as the end time of the trip;

[0068] D. Mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the first point of each segment as the starting city of the segment, and mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the last point of each segment as the ending city of the segment;

[0069] E. Calculate the spherical distance between every two adjacent points using the semi-versus formula. The sum of these values ​​is recorded as the distance traveled on the journey, where R is the Earth's radius.

[0070]

[0071]

[0072]

[0073]

[0074]

[0075] F. Subtract the start time of the trip from the end time of the trip to record the travel time.

[0076] Third: Examine the trip data and filter out low-quality trips.

[0077] In step three, low-quality trips are filtered according to the following rules:

[0078] A. Filter out trips with fewer than 10 satellite points;

[0079] B. Filter out trips with a mileage of less than 300 meters;

[0080] C. Filter trips with a travel time of less than 10 minutes.

[0081] Fourth: Statistically filter the cities where the starting and ending points of the trip are located.

[0082] In step four: 1) Given a truck, the route used is the route filtered in step three; 2) Calculate the prefecture-level cities where the start and end points of all the truck's routes are located after filtering, and calculate the average longitude of the satellite positioning of points belonging to the same prefecture-level city as the longitude of the truck's main stop point in this prefecture-level city, and calculate the average latitude of the satellite positioning of points belonging to the same prefecture-level city as the latitude of the truck's main stop point in this prefecture-level city.

[0083] Fifth: Calculate the distance between any two cities you have visited.

[0084] In step five, given a long-distance truck, calculate the distance between any two different prefecture-level cities where the truck stops. The specific steps for calculating the distance between two different prefecture-level cities are as follows:

[0085] A. Select two different prefecture-level cities from the cities of stay counted in step four;

[0086] B. The average value of the satellite positioning latitude and longitude calculated in step four represents the latitude and longitude coordinates of the prefecture-level city. That is, the calculated average value of the satellite positioning represents the center of the truck's stop in the prefecture-level city.

[0087] C. The semi-sine formula is used to calculate the spherical distance between two cities, thus calculating the distance between the two locations.

[0088] Sixth: Construct an intercity network flow for the travel trajectory of each vehicle.

[0089] In step six, given the travel data of a long-haul truck, the network flow of the truck is constructed, all possible routes are searched, and the flow of each route is calculated. The specific steps are as follows:

[0090] 1) Record each prefecture-level city counted in step four as a node in the network flow, arc This indicates that the truck's journey included a starting point in a prefecture-level city. The destination is a prefecture-level city. The itinerary, for The capacity indicates that the starting point is a prefecture-level city. The destination is a prefecture-level city. Number of trips;

[0091] 2) Path search: Given any two distinct nodes in a network flow and Let's assume Starting from a prefecture-level city, For the destination prefecture-level city, search for nodes in the network flow. To the node All paths, where each path does not contain duplicate nodes;

[0092] 3) The number of filtered path nodes exceeds the threshold ( The excessively long path from the prefecture-level city is obtained. to prefecture-level cities All path sets ;

[0093] 4) Calculation path Maximum capacity Its value is the minimum radian capacity of all radians in the path.

[0094] .

[0095] 5) Calculate the results for any two different prefecture-level cities. and All capacities between

[0096]

[0097]

[0098] .

[0099] Step 7: Sort algorithm for frequently used routes.

[0100] In step seven, all paths are sorted, and the specific steps are as follows:

[0101] 1) Based on the path length calculated in step five, sort all possible routes in descending order of distance from the start and end points, and select the route with the longest distance. The route, the result is ;

[0102] 2) Based on the path capacity calculated in step six, In The routes are sorted in descending order, and the result is: ;

[0103] 3) According to The result determines the threshold for path capacity. ,choose Routes with a capacity greater than this threshold are reordered in descending order of distance between the start and end points, resulting in: ;

[0104] 4) Select The front of the middle The route serves as the long-haul truck's... A frequently used route.

[0105] Example 2

[0106] like Figures 1 to 3 As shown: A method for identifying frequently used routes of long-haul trucks based on GPS data, comprising the following steps:

[0107] Step S10: Obtain GPS satellite positioning data.

[0108] The GPS satellite positioning data refers to data collected by vehicle-to-everything (V2X) data acquisition equipment, including pre-installed devices like T-Box, post-installed devices like OBD, rearview mirrors, and dashcams. This data includes satellite positioning longitude, latitude, time, system device time, and satellite accuracy. Common data point acquisition frequencies are 1 second, 15 seconds, or 30 seconds.

[0109] Step S20: Satellite data routing processing.

[0110] Step S2001, given a truck All collected GPS data points Sort by satellite positioning time in ascending order, with satellite positioning time accurate to year, month, day, hour, minute, and second.

[0111] Step S2002: Calculate the system equipment time interval between each collection point and the previous point. If the interval is greater than 10 minutes, then the point is regarded as the starting point of the next segment of the journey.

[0112] Step S2003: Mark the satellite positioning time of the first point of each trip segment as the start time of the trip, and mark the satellite positioning time of the last point of each trip segment as the end time of the trip.

[0113] Step S2004: Mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the first point of each trip as the departure city of the trip, and mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the last point of each trip as the destination city of the trip.

[0114] Step S2005: Based on the latitude and longitude of the starting and ending points of each historical journey, and the Earth's radius... m, the distance traveled in each segment of the journey is calculated using the semi-sine formula.

[0115] Step S30: Validate the travel data and filter out low-quality travels.

[0116] Step S3001: Filter historical routes whose starting point and ending point belong to the same prefecture-level city.

[0117] Step S3002: Filter out trips with fewer than 10 satellite points.

[0118] Step S3003: Filter out trips with a mileage of less than 300 meters.

[0119] Step S3004: Filter out trips with a travel time of less than 10 minutes.

[0120] Step S40: Statistically analyze the prefecture-level cities where the start and end points of the filtered trips are located.

[0121] Step S4001, given a truck The system will compile statistics on the origin and destination cities of all historical trips of the truck and store the results in a dataset. middle.

[0122] Step S4002, given a truck For any prefecture-level city Calculate trucks In prefecture-level cities The average longitude of all starting and ending points, denoted as the truck's longitude. In prefecture-level cities Longitude of the center of stay .

[0123] Step S4003, given a truck For any prefecture-level city Calculate trucks In prefecture-level cities The average latitude of all starting and ending points, denoted as the truck's latitude. In prefecture-level cities Latitude of the center of stay .

[0124] Step S50: Calculate the distance between any two cities where you have stayed.

[0125] Step S5001, given a truck For any two prefecture-level cities , ,in and According to Earth's radius ,Will , , , The formula for calculating the semi-versus is applicable to trucks. of , Distance between the two Earth surfaces .

[0126] Step S60: Construct intercity network flow .

[0127] Step S6001, given a truck Constructing directed network flows between cities ,like Figure 2 As shown, the vertices are a set. Each prefecture-level city in the arc Represents truck From prefecture-level cities , Number of times the trip has been recorded in history.

[0128] Step S6002, given a truck For any prefecture-level city , ,in and The algorithm for searching all possible routes between the two locations is as follows:

[0129] Step S600201: Construct a network stream in JSON format as described in step S60. The key is the prefecture-level city. The value is for trucks From prefecture-level cities Starting point: List of prefecture-level cities reachable in one trip. .

[0130] Step S600202, as follows Figure 3 As shown, the initial value of the parameter is set. An empty list used to store and output different route options. An empty list is used to store the prefecture-level cities traversed in a single path in sequence.

[0131] Step S6003, given a truck For any prefecture-level city , ,in and The algorithm for calculating the maximum capacity between two locations is as follows:

[0132] Step S600301, call Figure 3 In The algorithm will connect the two prefecture-level cities. , The route plan result is denoted as .

[0133] Step S600302, Filter out complex schemes, assigning a capacity value of 0 to schemes with more than 4 nodes, otherwise assigning a value that is not specified.

[0134] for:

[0135]

[0136] Step S600303: Considering the missing trips in the uploaded historical data, calculate... arrive Maximum capacity :

[0137]

[0138] Step S600304: Repeat the above three steps to calculate. arrive Maximum capacity .

[0139] Step S600305, will and The and the notation , used to calculate route capacity.

[0140] Step S70: Sort algorithm for frequently used routes.

[0141] Step S7001, given a truck For any prefecture-level city , ,in and Calculate the distance using step 4. Calculate the route capacity using step 5. .

[0142] Step S7002: Sort the routes by distance from origin to destination cities in descending order and select the top 20 routes with a capacity greater than 0. Save the results. .

[0143] Step S7003, will The routes are categorized by route capacity. Reordered, the result is .

[0144] Step S7004: Find the critical threshold of capacity. .remember The second largest capacity in the country is ,remember , If it does not exist, record it. .

[0145] Step S7005, filter out Medium capacity greater than The routes are sorted in descending order by the distance between the origin and destination cities.

[0146] Step S7006, The top 5 are used as the output of the final frequently used route.

[0147] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0148] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for identifying frequently traveled routes of trucks based on GPS data, characterized in that, Includes the following steps: Step 1: Acquire vehicle network data by using satellite positioning data provided by vehicle network acquisition equipment; Step 2: Data route conversion, which uses algorithms to merge discrete GPS points into multiple route segments; Step 3: Examine the trip data and filter out low-quality trips; Step 4: Statistically filter the cities where the origin and destination of the trip are located; Step 5: Calculate the distance between any two cities you have visited; Step 6: Construct an intercity network flow for the travel trajectory of each vehicle; Step 7: Sort algorithm for frequently used routes; In step six, given the travel data of a long-haul truck, the network flow of the truck is constructed, all possible routes are searched, and the flow of each route is calculated. The specific steps are as follows: 1) Record each prefecture-level city counted in step four as a node in the network flow, arc This indicates that the truck's journey included a starting point in a prefecture-level city. The destination is a prefecture-level city. The itinerary, for The capacity indicates that the starting point is a prefecture-level city. The destination is a prefecture-level city. Number of trips; 2) Path search: Given any two distinct nodes in a network flow and Let's assume Starting from a prefecture-level city, For the destination prefecture-level city, search for nodes in the network flow. To the node All paths, where each path does not contain duplicate nodes; 3) The number of filtered path nodes exceeds the threshold ( The excessively long path from the prefecture-level city is obtained. to prefecture-level cities All paths ;; 4) Calculation path Maximum capacity Its value is the minimum radian capacity of all radians in the path; ； 5) Calculate the results for any two different prefecture-level cities. and All capacities between ; ； ； ； In step seven, all paths are sorted, and the specific steps are as follows: 1) Based on the path length calculated in step five, sort all possible routes in descending order of distance from the start and end points, and select the route with the longest distance. The route, the result is ; 2) Based on the path capacity calculated in step six, In The routes are sorted in descending order, and the result is: ; 3) According to The result determines the threshold for path capacity. ,choose Routes with a capacity greater than this threshold are reordered in descending order of distance between the start and end points, resulting in: ; 4) Select The front of the middle The route serves as the long-haul truck's... A frequently used running route.

2. The method for identifying frequently used truck routes based on GPS data according to claim 1, characterized in that, In step one: the vehicle network data collection equipment includes a pre-installed T-Box, a rear-end data collection OBD device, a rearview mirror, and a driving recorder; the satellite positioning data includes satellite positioning longitude, satellite positioning latitude, satellite positioning time, system equipment time, and satellite accuracy.

3. The method for identifying frequently traveled routes of trucks based on GPS data according to claim 2, characterized in that, In step two, the data runout processing includes: A. Given all the GPS data points collected for a truck, sort them in ascending order by satellite positioning time; B. Calculate the system equipment time interval between each collection point and the previous point. If the interval is greater than 10 minutes, then the point is regarded as the starting point of the next segment of the journey. C. Mark the satellite positioning time of the first point of each trip as the start time of the trip, and mark the satellite positioning time of the last point of each trip as the end time of the trip; D. Mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the first point of each segment as the starting city of the segment, and mark the prefecture-level city corresponding to the satellite positioning latitude and longitude of the last point of each segment as the ending city of the segment; E. Calculate the spherical distance between every two adjacent points using the semi-versus formula. The summation is recorded as the distance traveled on the trip, where R is the Earth's radius; ； ； ； ； ； F. Subtract the start time of the trip from the end time of the trip to record the travel time.

4. The method for identifying frequently used truck routes based on GPS data according to claim 1, characterized in that, In step three, low-quality trips are filtered according to the following rules: A. Filter out trips with fewer than 10 satellite points; B. Filter out trips with a mileage of less than 300 meters; C. Filter trips with a travel time of less than 10 minutes.

5. The method for identifying frequently traveled routes of trucks based on GPS data according to claim 1, characterized in that, In step four: 1) Given a truck, the route used is the route filtered in step three; 2) After statistical filtering, calculate the prefecture-level cities where the starting and ending points of all the truck's trips are located, and record the average longitude of the points belonging to the same prefecture-level city as the longitude of the main stop of the truck in that prefecture-level city. Calculate the average latitude of the points belonging to the same prefecture-level city as the latitude of the main stop of the truck in that prefecture-level city.

6. The method for identifying frequently traveled routes of trucks based on GPS data according to claim 5, characterized in that, In step five, given a long-distance truck, calculate the distance between any two different prefecture-level cities where the truck stops. The specific steps for calculating the distance between two different prefecture-level cities are as follows: A. Select two different prefecture-level cities from the cities of stay counted in step four; B. The average value of the satellite positioning latitude and longitude calculated in step four represents the latitude and longitude coordinates of the prefecture-level city. That is, the calculated average value of the satellite positioning represents the center of the truck's stop in the prefecture-level city. C. The semi-sine formula is used to calculate the spherical distance between two cities, thus calculating the distance between the two locations.

Images