Real-time taxi cruising route recommendation method based on big data
Through the real-time taxi cruise route recommendation system based on big data, taxi drivers have solved the problems of low operational efficiency and difficulty in carrying passengers caused by taxi drivers relying on intuition and experience to plan routes, and achieved more efficient route planning and carrying passengers.
Patent Information
- Application Number
- CN202010451752.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-09-10
- Publication Date
- 2025-05-02
- Estimated Expiration
- 2039-09-10
AI Technical Summary
Taxi drivers traditionally rely on intuition and experience to plan cruising routes, resulting in low operational efficiency and difficulty in carrying passengers efficiently.
A taxi real-time cruise route recommendation system based on big data is adopted, and the candidate route acquisition module and real-time cruise route selection module are obtained and evaluated multiple candidate routes, and routes with high crowd density index or existing passenger transport centers are selected as recommended routes.
It improves the operational efficiency and passenger frequency of taxi drivers on the cruising route, ensuring that taxi can carry passengers and operate efficiently.
Smart Images
Figure CN111798026B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of big data, and in particular to a method for recommending real-time taxi cruising routes based on big data. Background Art
[0002] With the improvement of economic development, people's transportation needs are increasing. People not only need city buses and long-distance passenger transport, which are relatively common modes of transportation, but also require convenient, fast, safe and comfortable passenger transport. Taxis are popular because of their flexible and flexible operating characteristics. Taxis do not follow fixed routes, but the drivers plan their own cruising routes. The traditional practice of taxi drivers is to plan cruising routes based on intuition and experience. This method is sometimes very difficult to achieve efficient operation and passenger transportation. Summary of the invention
[0003] The purpose of the present invention is to provide a real-time taxi cruising route recommendation system and method based on big data to solve the problems in the prior art.
[0004] To achieve the above object, the present invention provides the following technical solutions:
[0005] A taxi real-time cruising route recommendation system based on big data, the recommendation system comprises a candidate route acquisition module and a real-time cruising route selection module, the candidate route acquisition module is used to obtain multiple candidate routes for taxi real-time cruising, and the real-time cruising route is used to select a real-time cruising recommended route from the candidate routes.
[0006] More optimally, the candidate route acquisition module includes a starting position reading module, a target position input module and a candidate route generation module. The starting position reading module is used to read the current position of the taxi, the target position input module is used for the driver to input the target position, and the candidate route generation module generates a number of candidate routes from the current position to the target position based on the current position and the target position of the taxi.
[0007] More optimally, the real-time cruising route selection module includes a passenger transport center search module, a crowd density index evaluation module and a real-time cruising recommended route determination module. The passenger transport center search module is used to search whether there is a passenger transport center on each candidate route, the crowd density index evaluation module is used to evaluate the crowd density index on each candidate route, and the real-time cruising recommended route determination module determines the real-time cruising recommended route based on the search results of the passenger transport center search module or the evaluation results of the crowd density index evaluation module.
[0008] More optimally, the crowd density index calculation module includes an interval distance acquisition module, an interval fluctuation calculation module, a search number acquisition module, a search number calculation module, a consumer number acquisition module, a consumer number calculation module, a route length acquisition module, a crowd density index calculation module and a candidate route sorting module. The interval distance acquisition module is used to acquire the distance between adjacent merchants on each candidate route, the interval fluctuation calculation module is used to calculate the degree of fluctuation of the distance between adjacent merchants on each candidate route, the search number acquisition module is used to acquire the number of searches on the server for merchants on each candidate route in the first target time period, the search number calculation module is used to calculate the average number of searches on the server for each merchant on a candidate route, the consumer number acquisition module is used to acquire the number of consumers of each merchant on each candidate route in a certain time period every day in the second target time period, the consumer number calculation module is used to calculate the average number of consumers of all merchants on a certain candidate route in a certain time period, the route length acquisition module is used to acquire the length of each candidate route, the crowd density index calculation module calculates the crowd density index of a certain candidate route, and the candidate route sorting module is used to perform sorting evaluation according to the crowd density index of each candidate route.
[0009] A method for recommending real-time taxi cruising routes based on big data, the method comprising the following steps:
[0010] S1: Obtain candidate routes;
[0011] S2: Select the real-time cruising recommended route of the taxi from the candidate routes.
[0012] Preferably, the recommendation method further comprises the following steps:
[0013] S1: Read the current location of the taxi, the driver enters the target location, and obtains all candidate routes A1, A2, A3, ...A from the current location of the taxi to the target location n , n is the number of candidate routes;
[0014] S2: Search whether there is a passenger transport center on each candidate route. If there is a passenger transport center on the candidate route, the real-time cruising recommended route for the taxi is the candidate route. If there is no passenger transport center on the candidate route, the crowd density index of each candidate route is evaluated, and the real-time cruising recommended route for the taxi is selected according to the crowd density index of each candidate route.
[0015] Preferably, the step S2 of evaluating the crowd density index of each candidate route includes the following steps:
[0016] S21: Obtain the distance A between adjacent businesses on each candidate route 11=[B 11 , B 12 , B 13 , …, B 1(m1-1) ]、A 21 =[B 21 , B 22 , B 23 , …, B 2(m2-1) ]、A 31 =[B 31 , B 32 , B 33 , …, B 3(m3-1) ], …, A i1 =[B i1 , B i2 , B i3 , …, B ij , …, B i(mi-1) ] , …, A n1 =[B n1 , B n2 , B n3 , …, B n(mn-1) ], where A i1 is the set of distances between adjacent businesses on the i-th candidate route, B ij is the distance between the jth merchant and the j+1th merchant on the ith candidate route, and m1, m2, m3, …, mi, …mn are the candidate routes A1, A2, A3, …, A i , …, A n The number of merchants on the Internet, 1<=i<=n,
[0017] For each candidate route, the fluctuation degree of the distance between adjacent businesses on the i-th candidate route is calculated using the following formula:
[0018] , ,
[0019] Among them, L i is the route distance between the first merchant and the last merchant in the direction from the current location of the taxi to the target location on the i-th candidate route, E i is the fluctuation degree of the distance between adjacent merchants on the i-th candidate route, and mi is the total number of merchants on the i-th candidate route. The smaller the fluctuation degree between adjacent merchants, the more evenly the merchants are distributed, and the more stable the probability of merchants picking up passengers.
[0020] S22: Obtain the number of times A that a merchant on each candidate route has been searched on the server in the first target time period T1. 12 =[C 11 , C 12 , C 13,…,C 1m1 ]、A 22 =[C 21 , C 22 , C 23 ,…,C 2m2 ]、A 32 =[C 31 , C 32 , C 33 ,…,C 3m3 ], …, A i2 =[C i1 , C i2 , C i3 ,…,C ij ,…,C imi ] , …, A n2 =[C n1 , C n2 , C n3 ,…,C nmn ], where A i2 is the collection of the number of searches for each merchant on the i-th candidate route on the server, C ij is the number of times the j-th merchant on the i-th candidate route is searched on the server;
[0021] For each candidate route, the following formula is used to calculate the average number of times each merchant is searched on the server on the i-th candidate route:
[0022] ; If a merchant is searched more times on the server, the merchant has more potential customers, so the merchant has more potential passengers on the candidate routes.
[0023] S23: Obtain the number of consumers of each merchant on each candidate route in a certain time period T3 of each day in the second target time period T2, and calculate the average number of consumers of all merchants on the candidate route in a certain time period T3 using the following formula:
[0024] X i = ;
[0025] Among them, X i represents the average number of customers at all merchants on the i-th candidate route in a certain period of time T3, P jk is the number of consumers at the jth merchant on the i-th candidate route in a certain time period T3 on the kth day in the second target time period T2, m i is the number of merchants on the i-th candidate route, T2 represents the number of days in the second target time period; the more average consumption times, the more traffic there is on the candidate route, and it is easier for taxi drivers to pick up passengers in places with high traffic.
[0026] S24: Get the length S1, S2, S3, ... S of each candidate route respectively n , S i is the length of the i-th candidate route. For each candidate route, the crowd density index of the candidate route is calculated using the following calculation formula:
[0027] ,
[0028] Among them, Y i is the crowd density index of the i-th candidate route, H i is the average number of times each merchant on the i-th candidate route is searched on the server, E i is the fluctuation degree of the distance between adjacent merchants on the i-th candidate route, X i is the average number of consumers of all merchants on the i-th candidate route in a certain period of time T3. By considering the crowd density index from multiple angles, it is conducive to a more objective evaluation of the crowd density index of the candidate route.
[0029] S25: Sort the crowd density indexes of all candidate routes in descending order, and take the candidate route corresponding to the highest density index as the real-time cruise recommended route. The higher the crowd density index, the greater the probability of the taxi picking up passengers, and the higher the frequency of picking up passengers.
[0030] More optimally, the passenger transport center includes a bus station, a railway station and a high-speed railway station.
[0031] Compared with the prior art, the beneficial effects of the present invention are: the present invention obtains the starting position of the taxi, the driver's target position input module obtains all candidate routes, and then selects the real-time cruising recommended route from the candidate routes, so that the taxi driver can efficiently operate and carry passengers on the cruising recommended route. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 This is a module schematic diagram of a taxi real-time cruising route recommendation system based on big data in the present invention;
[0033] Figure 2 The present invention is a flowchart of a method for recommending real-time taxi cruising routes based on big data. DETAILED DESCRIPTION
[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 are within the scope of protection of the present invention.
[0035] See also Figures 1-2 In an embodiment of the present invention, a taxi real-time cruising route recommendation system based on big data, the recommendation system includes a candidate route acquisition module and a real-time cruising route selection module, the candidate route acquisition module is used to obtain multiple candidate routes for real-time cruising of taxis, and the real-time cruising route is used to select a real-time cruising recommended route from the candidate routes.
[0036] The candidate route acquisition module includes a starting position reading module, a target position input module and a candidate route generation module. The starting position reading module is used to read the current position of the taxi, the target position input module is used for the driver to input the target position, and the candidate route generation module generates several candidate routes from the current position to the target position according to the current position and the target position of the taxi.
[0037] The real-time cruise route selection module includes a passenger transport center search module, a crowd density index evaluation module and a real-time cruise recommended route determination module. The passenger transport center search module is used to search whether there is a passenger transport center on each candidate route. The crowd density index evaluation module is used to evaluate the crowd density index on each candidate route. The real-time cruise recommended route determination module determines the real-time cruise recommended route based on the search results of the passenger transport center search module or the evaluation results of the crowd density index evaluation module.
[0038] The crowd density index calculation module includes an interval distance acquisition module, an interval fluctuation calculation module, a search number acquisition module, a search number calculation module, a consumer number acquisition module, a consumer number calculation module, a route length acquisition module, a crowd density index calculation module and a candidate route sorting module. The interval distance acquisition module is used to obtain the distance between adjacent businesses on each candidate route, the interval fluctuation calculation module is used to calculate the fluctuation degree of the distance between adjacent businesses on each candidate route, the search number acquisition module is used to obtain the number of searches on the server for businesses on each candidate route in the first target time period, the search number calculation module is used to calculate the average number of searches on the server for each business on a candidate route, the consumer number acquisition module is used to obtain the number of consumers of each business on each candidate route in a certain time period every day in the second target time period, the consumer number calculation module is used to calculate the average number of consumers of all businesses on a certain candidate route in a certain time period, the route length acquisition module is used to obtain the length of each candidate route, the crowd density index calculation module calculates the crowd density index of a certain candidate route, and the candidate route sorting module is used to sort and evaluate according to the crowd density index of each candidate route.
[0039] A method for recommending real-time taxi cruising routes based on big data, the recommendation method comprising the following steps:
[0040] S1: The starting position reading module reads the current location of the taxi, and the driver enters the target location in the target position input module, and then obtains all candidate routes A1, A2, A3, ... A from the current location of the taxi to the target location from the candidate route generation module. n , n is the number of candidate routes; in the actual acquisition of candidate routes, the length of the longest candidate route can be limited to be less than or equal to twice the length of the shortest candidate route, so as to prevent the generation of too many candidates and too much impurity.
[0041] S2: The passenger transport center search module searches whether there is a passenger transport center on each candidate route. The passenger transport center includes bus stations, railway stations and high-speed railway stations. If there is a passenger transport center on the candidate route, the real-time cruising recommended route determination module determines that the real-time cruising recommended route of the taxi is the candidate route. If there is no passenger transport center on the candidate route, the crowd density index evaluation module evaluates the crowd density index of each candidate route. The real-time cruising recommended route determination module selects the real-time cruising recommended route of the taxi according to the crowd density index of each candidate route.
[0042] The evaluation of the crowd density index of each candidate route in step S2 includes the following steps:
[0043] S21: The interval distance acquisition module obtains the distance A between adjacent businesses on each candidate route. 11 =[B 11 , B 12 , B 13 , …, B 1(m1-1) ]、A 21 =[B 21 , B 22 , B 23 , …, B 2(m2-1) ]、A 31 =[B 31 , B 32 , B 33 , …, B 3(m3-1) ], …, A i1 =[B i1 , B i2 , B i3 , …, B ij , …, B i(mi-1) ] , …, A n1 =[B n1 , B n2 , B n3 , …, B n(mn-1) ], where A i1 is the set of distances between adjacent businesses on the i-th candidate route, B ijis the distance between the jth merchant and the j+1th merchant on the ith candidate route, and m1, m2, m3, …, mi, …mn are the candidate routes A1, A2, A3, …, A i , …, A n The number of merchants on the Internet, 1<=i<=n,
[0044] For each candidate route, the interval fluctuation calculation module calculates the fluctuation degree of the distance between adjacent merchants on the i-th candidate route using the following formula:
[0045] , ,
[0046] Among them, L i is the route distance between the first merchant and the last merchant in the direction from the current location of the taxi to the target location on the i-th candidate route, E i is the fluctuation degree of the distance between adjacent merchants on the ith candidate route, and mi is the total number of merchants on the ith candidate route;
[0047] S22: The search count acquisition module acquires the number of searches A of the merchants on each candidate route on the server in the previous seven days. 12 =[C 11 , C 12 , C 13 ,…,C 1m1 ]、A 22 =[C 21 , C 22 , C 23 ,…,C 2m2 ]、A 32 =[C 31 , C 32 , C 33 ,…,C 3m3 ], …, A i2 =[C i1 , C i2 , C i3 ,…,C ij ,…,C imi ] , …, A n2 =[C n1 , C n2 , C n3 ,…,C nmn ], where A i2 is the collection of the number of searches for each merchant on the i-th candidate route on the server, C ij is the number of times the j-th merchant on the i-th candidate route is searched on the server;
[0048] For each candidate route, the search count calculation module uses the following formula to calculate the average number of times each merchant is searched on the server on the i-th candidate route:
[0049] ;
[0050] S23: The consumer number acquisition module obtains the number of consumers from each merchant on each candidate route during the period from 9:00 to 15:00 every day in the previous ten days. The consumer number calculation module calculates the average number of consumers from 9:00 to 15:00 for all merchants on the candidate route using the following formula:
[0051] X i = ;
[0052] Among them, X i represents the average number of customers from 9:00 to 15:00 in the first ten days at all merchants on the i-th candidate route, P jk is the number of consumers at the jth merchant on the i-th candidate route in a certain time period T3 on the kth day in the first ten days, m i is the number of businesses on the i-th candidate route. A certain time period T3 can be consistent with the driver's cruising time. For example, if a taxi driver plans to cruise between midnight and 4 a.m. the next day, then the certain time period T3 is from midnight to 4 a.m. the next day.
[0053] S24: The route length acquisition module acquires the length S1, S2, S3, ... S of each candidate route respectively. n , S i is the length of the i-th candidate route. For each candidate route, the crowd density index calculation module calculates the crowd density index of the candidate route using the following calculation formula:
[0054] ,
[0055] Among them, Y i is the crowd density index of the i-th candidate route, H i is the average number of times each merchant on the i-th candidate route is searched on the server, E i is the fluctuation degree of the distance between adjacent merchants on the i-th candidate route, X i is the average number of consumers at all merchants on the i-th candidate route in a certain period of time T3.
[0056] S25: The candidate route sorting module sorts the crowd density indexes of all candidate routes in descending order, and selects the candidate route corresponding to the highest crowd density index as the real-time cruise recommended route. The higher the crowd density index of a candidate route, the greater the probability that a taxi will pick up passengers on the candidate route, and the higher the frequency of picking up passengers.
[0057] It will be apparent to those skilled in the art that the 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 the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. A real-time taxi cruising route recommendation method based on big data, characterized by: The recommended method comprises the following steps: S1: Obtain candidate routes; S2: Select the real-time cruising recommended route of the taxi from the candidate routes; The recommended method further comprises the following steps: S1: Read the current location of the taxi, the driver inputs the target location, and obtains all candidate routes A1, A2, A3, ...An from the current location of the taxi to the target location, where n is the number of candidate routes; S2: Search whether there is a passenger transport center on each candidate route. If there is a passenger transport center on the candidate route, the real-time cruising recommended route for taxis is the candidate route. If there is no passenger transport center on the candidate route, the crowd density index of each candidate route is evaluated, and the real-time cruising recommended route for taxis is selected according to the crowd density index of each candidate route. The step S2 of evaluating the crowd density index of each candidate route includes the following steps: S21: Obtain the distance A between adjacent businesses on each candidate route 11 =[B 11 , B 12 , B 13 , …, B 1(m1-1) ]、A21=[B 21 , B 22 , B 23 , …, B 2(m2-1) ]、A 31 =[B 31 , B 32 , B 33 , …, B 3(m3-1) ], …, A i1 =[B i1 , B i2 , B i3 , …, B ij , …, B i(mi-1) ], …, A n1 =[B n1 , B n2 , B n3 , …, B n(mn-1) ], where A i1 is the set of distances between adjacent businesses on the i-th candidate route, B ij is the distance between the jth merchant and the j+1th merchant on the ith candidate route, and m1, m2, m3, …, mi, …mn are the candidate routes A1, A2, A3, …, A i , …, A n The number of merchants on the Internet, 1<=i<=n, For each candidate route, the interval fluctuation calculation module calculates the fluctuation degree of the distance between adjacent merchants on the i-th candidate route using the following formula: Among them, L i is the route distance between the first and last merchants in the direction from the current location of the taxi to the target location on the ith candidate route, Ei is the fluctuation degree of the distance between adjacent merchants on the ith candidate route, and mi is the total number of merchants on the ith candidate route; S22: Obtain the number of times A that a merchant on each candidate route has been searched on the server in the first target time period T1. 12 =[C 11 , C 12 , C 13 ,…,C 1m1 ]、A 22 =[C 21 , C 22 , C 23 ,…,C 2m2 ]、A 32 =[C 31 , C 32 , C 33 ,…,C 3m3 ], …, A i2 =[C i1 , C i2 , C i3 ,…,C ij ,…,C imi ], …, A n2 =[C n1 , C n2 , C n3 ,…,C nmn ], where A i2 is the collection of the number of searches for each merchant on the i-th candidate route on the server, C ij is the number of times the j-th merchant on the i-th candidate route is searched on the server; For each candidate route, the search count calculation module uses the following formula to calculate the average number of times each merchant is searched on the server on the i-th candidate route: S23: Obtain the number of consumers of each merchant on each candidate route in a certain time period T3 of each day in the second target time period T2, and calculate the average number of consumers of all merchants on the candidate route in a certain time period T3 using the following formula: Among them, X i represents the average number of customers at all merchants on the i-th candidate route in a certain period of time T3, P jk is the number of consumers of the j-th merchant on the i-th candidate route in a certain time period T3 on the k-th day in the second target time period T2, mi is the number of merchants on the i-th candidate route, T2 represents the number of days in the second target time period, S24: Obtain the length S1, S2, S3, ... S of each candidate route respectively n , S i is the length of the i-th candidate route. For each candidate route, the crowd density index of the candidate route is calculated using the following calculation formula: Among them, Y i is the crowd density index of the i-th candidate route, H i is the average number of times each merchant on the i-th candidate route is searched on the server, E i is the fluctuation degree of the distance between adjacent merchants on the i-th candidate route, X i is the average number of customers at all merchants on the i-th candidate route in a certain period of time T3; S25: The candidate route sorting module sorts the crowd density indexes of all candidate routes in descending order, and takes the candidate route corresponding to the highest crowd density index as the real-time cruise recommended route; The method is applied to a system in which the recommendation system comprises a candidate route acquisition module and a real-time cruising route selection module, wherein the candidate route acquisition module is used to acquire candidate routes of multiple taxi real-time cruising routes, and the real-time cruising route is used to select a real-time cruising recommended route from the candidate routes.