CN110245377B - Travel scheme recommendation method and recommendation system - Google Patents
Travel scheme recommendation method and recommendation system Download PDFInfo
- Publication number
- CN110245377B CN110245377B CN201910378683.7A CN201910378683A CN110245377B CN 110245377 B CN110245377 B CN 110245377B CN 201910378683 A CN201910378683 A CN 201910378683A CN 110245377 B CN110245377 B CN 110245377B
- Authority
- CN
- China
- Prior art keywords
- time
- carriage
- degree
- station
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 239000011159 matrix material Substances 0.000 claims abstract description 28
- 230000003993 interaction Effects 0.000 claims abstract description 6
- 238000005192 partition Methods 0.000 claims description 21
- 241001123248 Arma Species 0.000 claims description 19
- 238000001914 filtration Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000010586 diagram Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000007689 inspection Methods 0.000 claims description 6
- 238000012821 model calculation Methods 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 6
- 230000000739 chaotic effect Effects 0.000 claims description 5
- 238000012546 transfer Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 2
- 239000002131 composite material Substances 0.000 claims 1
- 238000012545 processing Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/123—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a travel scheme recommending method, which comprises the following steps: receiving inquiry information of a user; predicting the number of waiting vehicles at the platform; predicting the degree of congestion of the carriage; predicting the arrival time of a vehicle; and (3) judging an optimal scheme: and (3) integrating the attributes of the number of waiting vehicles at the platform, the degree of congestion of the carriage and the arrival time, establishing a standardized decision matrix, assigning weights to the attributes, calculating a weighted standardized decision matrix, and giving out the recommendation of the optimal travel scheme after matching the user inquiry information. The invention also discloses a travel scheme recommendation system, which comprises an interaction module, a platform waiting number judging module, a crowdedness judging module, a arrival time judging module and a travel scheme judging module. The invention comprehensively considers the crowding degree of the carriage and improves the traveling quality of passengers.
Description
Technical Field
The invention relates to the technical field of video image processing and the technical field of mathematical modeling, in particular to a travel scheme recommendation method and a travel scheme recommendation system based on digital image processing and multi-objective decision.
Background
With the continuous expansion of urban scale and population, the internet and the internet of things becoming mature, our daily life is toward the aspects of intellectualization, informatization and the like, the concept of "smart city" becomes the main stream direction of urban construction, and the vigorous rise of intelligent traffic systems not only provides convenience for everyone, but also becomes the best way for solving traffic problems currently internationally acknowledged. At present, a plurality of real-time public buses and APP are sequentially pushed out, and the intelligent buses are also indicated to meet the requirements of people in the large development direction in the future while the market is preempted, so that the problem of 'pain spot' of the existing buses is solved. In regions such as Shandong and Zhejiang in China, the intelligent bus station is tried to provide information service for passengers to travel by means of cloud computing and mass data processing capability of a cloud platform.
Today, the increasing demand of intelligent public transportation is how to provide optimal routes according to real-time traffic conditions and user inquiry information, which is an important problem that we have to think about. In the aspect of intelligent management of urban public transportation, how to improve the satisfaction of passengers and the riding experience is the core content of intelligent public transportation system research.
Current public transportation APP such as a God map, a hundred degree map and the like displays a public transportation trip scheme including route information (direct and transfer), time information (arrival time and total time) and the like. Patent CN201810858818.5 of Chongqing city comprehensive transportation hub (group) limited company and Chongqing transportation university discloses a public transportation travel guidance system, and patent CN201610193300.5 of southwest transportation university provides an integrated travel comprehensive decision support model, but the riding experience is not enough, the influence of the crowding degree of the carriage on the riding experience is not considered, for example, passengers with physical disabilities or traveling with large pieces of luggage tend to wait for a vehicle with low crowding degree except riding time.
In addition, the prediction or monitoring of the traffic is mostly based on a macroscopic angle, namely, the total traffic within a certain range is detected for a certain time, so that the traffic is regulated and controlled on the macroscopic angle. Such as people flow monitoring in the scenes of a large mall, a railway station, a bus stop, a busy road and the like. Because the pointing range is wider, the pertinence is not strong for common users, and the information which can be extracted from the pointing range is very limited.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art and provides a travel scheme recommending method and a travel scheme recommending system.
The aim of the invention is achieved by the following technical scheme: a travel scheme recommending method comprises the following steps:
receiving inquiry information of a user;
predicting the number of waiting persons at a platform: acquiring time sequences of the number of passengers getting on and off the same route, the same platform and the same time period, and predicting the number of passengers waiting at the platform;
predicting the degree of congestion of the carriage: the number of waiting vehicles at the platform and the number of passengers in the carriage are combined for statistics and prediction of the degree of carriage crowding;
predicting the arrival time of a vehicle;
and (3) judging an optimal scheme: and (3) integrating the attributes of the number of waiting vehicles at the platform, the degree of congestion of the carriage and the arrival time, establishing a standardized decision matrix, assigning weights to the attributes, calculating a weighted standardized decision matrix, and giving out the recommendation of the optimal travel scheme after matching the user inquiry information. According to the method, on the aspect of considering time factors, the carriage crowding degree is creatively added as a reference factor for proposal recommendation, and the combination of the number of passengers getting on and off a platform and the information quantity obtained by video in the middle of the carriage is proposed as a crowding degree judgment standard, so that the real-time crowding degree of each bus is accurately obtained, the requirements of more passengers can be met, and the method is more humanized.
Preferably, the step of predicting the number of bus stops is:
collecting the number of people getting on or off the bus through cameras of front and rear doors of the bus, and extracting the number of people getting on or off the bus at the same platform, the same line and the same time period after the data volume reaches a certain degree to form a number sequence;
after the stability of the platform is checked, a time sequence model is established, the ARMA model is utilized to predict the number of passengers on and off the platform, and meanwhile, reference factors such as weather, holidays and the like are added, so that the prediction result is more accurate.
Further, the modeling step of the ARMA model is as follows:
stability of the population sequence: generating a time sequence diagram by using the acquired data, judging the stability of the data by observing the time sequence diagram, wherein the stable sequence has the properties of constant mean and constant variance according to the definition of the stability, so that the time sequence diagram of the stable sequence fluctuates around a constant value, and the fluctuation range is bounded; meanwhile, according to actual conditions, the number of passengers getting on or off the platform is presumed not to be obviously increased or decreased in a short time;
model identification and grading: the ARMA (p, q) model is of the form:
wherein phi is 1 ，φ 2 ，…，φ p 、θ 1 ，θ 2 ，…，θ q And epsilon t ，ε t-1 ，…，ε 1 Respectively an autoregressive parameter, a moving average parameter and a white noise sequence, X t-1 ...X t-p The input independent variable sequence of the model is the history basis of prediction; preliminary judging the orders p and q of the ARMA model according to the autocorrelation coefficients and the partial autocorrelation coefficients, and calculating a sample Autocorrelation Coefficient (ACF) and a Partial Autocorrelation Coefficient (PACF);
parameter estimation of the model: according to the identified model and the order thereof, carrying out parameter estimation on the model by adopting a least square method;
and (3) model inspection: performing model identification, order determination and parameter estimation, and performing residual sequence inspection analysis on the model; the fitting model is only valid when the residual sequence is a white noise sequence, and can be used for prediction; s-step autocorrelation coefficient ρ of residual sequence { ε (t) } is calculated 1 、ρ 2 、...ρ s Then construct chi-square statistic F s F is shown in the following formula s Subject to degrees of freedom s beta 2 Distribution; wherein n is the capacity of the residual sequence;
predicting by using ARMA model calculation results; after inputting data such as lines, platforms, time and the like, the predicted waiting number can be obtained; and taking the ARMA model calculation result as a main prediction basis, and simultaneously adding reference factors such as weather, holidays and the like, so that the prediction result is more accurate.
Preferably, the step of predicting the car congestion degree includes:
extracting image partition HOG characteristics through a camera in the middle of the carriage, and pre-judging the crowding degree of the carriage;
capturing heads of people through a carriage door camera, counting the number of people getting on and off, and obtaining the number of people in the carriage by accumulating the number of people getting on and off each platform and subtracting the number of people getting off;
and calculating the arrival crowdedness degree by combining the pre-judging result and the number of the carriages.
The two crowding degree analyses of people counting and image analysis are mutually complemented, so that the accuracy of crowding degree judgment can be improved.
Furthermore, the extracting image partition HOG features, and pre-judging the crowding degree of the carriage specifically comprises:
extracting features of a directional gradient histogram (Histogram of Oriented Gradient, HOG) and calculating information quantity, extracting HOG features of an input video image in the middle of a carriage, setting a certain threshold value for each image partition, screening out an effective gradient value, counting an angle histogram of an effective gradient, dividing the image into partitions, taking g angle directions for each partition, and calculating gradient values of the g angles; variance G of combined effective gradient values 2 And the image information entropy H is calculated as follows:
h in the above formula represents the h-th angle direction in the g angle directions; mu is the average gradient value; p (P) ij A comprehensive feature representing gray values at a certain pixel position and gray distribution of pixels around the gray values; selecting the field gray average value of the image as a space characteristic value of gray distribution, and forming a characteristic binary group with the pixel gray of the image, wherein i represents the gray value of the pixel, i is more than or equal to 0 and less than or equal to 255, j represents the field gray value, and j is more than or equal to 0 and less than or equal to 255; f (i, j) is the frequency at which feature doublet (i, j) occurs; u, V are respectively the lengths of the imagesAnd width; the information entropy of the image is H;
calculating the confusion degree E= (G, H) of the final image pixels, and when the partition gradient value variance is small and the information entropy is large, indicating that the pixel value distribution of the partition is more chaotic, then the partition congestion degree is large; the statistics results of the number of waiting vehicles at the platform and the number of passengers in the carriage are mutually complemented with the judgment results of the crowding degree in the middle of the carriage, and the corresponding statistics value of the number of passengers is larger when the carriage is crowded, so that the final carriage crowding degree can be cooperatively judged by the statistics results of the number of passengers and the calculation results of the crowding degree of the middle camera, and the final carriage crowding degree can be used as the judgment basis of a travel scheme.
Further, the step of predicting the car congestion degree specifically includes:
calculating the real-time crowding degree: the method mainly judges according to the accumulated value of the number of people getting on and off the bus after the bus is taken, and corrects the result by utilizing the crowding degree calculated by the camera in the middle of the bus, wherein the crowding degree of the bus is judged in advance in an auxiliary mode by extracting the image partition HOG characteristic through the camera in the middle of the bus, and the method is calculated as follows:
wherein E is real_time The chaotic degree of the image pixels is queried in real time, namely the real-time crowding degree;is the accumulated value of the number of people getting on and off after departure; area is the Area that can stand; e (E) DIP The result of the image confusion degree judged by the camera in the middle of the carriage;
calculating the congestion degree of the arrival station: from real-time congestion level E real_time The station between the station where the inquiring vehicle is located and the passenger station is accumulated, the history statistics result of the number of passengers getting on and off the intermediate station is predicted, and the prediction of the arrival congestion degree can be realized, and the calculation is as follows:
wherein E is getin_station Is the degree of congestion of the arrival at the station,the number of passengers getting on or off the bus from the searched station to the station where the passenger is located.
Preferably, the step of predicting the arrival time of the vehicle includes:
calculating Euclidean distance between current data and historical data under the condition of original data matching by adopting a historical state matching method, so as to predict arrival time;
searching a large amount of historical data by the nonparametric regression method to find a set most similar to the current input state so as to predict the state at the next moment; assuming that the bus arrives at a k-1 station, and the data state corresponding to the starting time is S; by T km And T' km Respectively representing the inter-station travel time from k-1 station to k station and the stop time of k station corresponding to the m-th record with the data state of S in the historical database; taking J as the neighbor number of the non-parametric regression, and D as the dimension of the non-parametric regression; using Euclidean distance a m Matching the bus with historical data, selecting J neighbors, wherein A represents a set of the J neighbors, and predicting the travel time T of the bus from k stations to k+1 stations according to the neighbors by using a weighted average method k ：
e is a neighbor in a set of a different from m; similarly, the stop time of the bus at the station k can be predicted by using the formulas through the stop time of the stations k-d, …, k-2 and k-1, and d is the station number different from the station k.
Further, when the arrival time of the vehicle is predicted, the sum of the predicted stop time of the bus at the current station and the predicted running time of the bus at the next station is taken as the predicted running time and is taken as the input parameter of a Kalman filtering algorithm, so that the dynamic correction of a predicted result is realized;
the sum of the predicted stop time of the bus at the k-1 station and the predicted running time from the k-1 station to the k station is the predicted running time and is used as an input parameter of a Kalman filtering algorithm;
the Kalman filtering algorithm is to make an estimation meeting the minimum mean square error for the signal to be processed by the established state equation and the observation equation, complete the estimation of the state variable by combining the estimation value of the last moment and the observation value of the current moment, and solve the current estimation value; the state equation of the Kalman filter for bus arrival time prediction is as follows:
wherein t is (k+1) 、t' (k+1) The method comprises the steps of respectively representing the time and the observation time of a bus from a start station to a kth station, u (k) represents the inter-station travel time of the bus from k to k+1 stations, S (k), I (k) and M (k) are respectively data state transition variables, input variables and measured value coefficients, w (k) and v (k) are respectively input noise and measured noise, the state transition variables, the input variables and the measured value coefficients of Kalman filtering are all 1, and the input noise and the measured noise are independent white noise with the average value of 0;
when the bus reaches the kth station, kalman filtering is performed according to the observed value T of the kth station k m+1 And calculating the optimal estimated value of the kth site by the k-1 basic time sequences, further obtaining the adjustment value of the kth+1 site, sequentially updating the time of the subsequent site according to the adjustment value, and adding the updated time sequence into the basic time sequence.
Preferably, the step of determining the optimal scheme is as follows:
standardizing a decision matrix; let the travel scheme decision matrix be z= (Z xy ) L.W, wherein x is trip plan subscript, y plan attribute subscript, z xy Values representing the y attributes of the x schemes, L, W represent the length and width of the decision matrix, i.e. the total number of schemes and the total number of attributes of each scheme, respectively; the scheme attributes include the degree of congestion of the carriage,The number of waiting vehicles, the arrival time, the transfer times and the like. Let normalized decision matrix b= (B) xy ) L.W, where b xy Values representing the normalized x schema y attribute are:
x＝1,2...,L；y＝1,2...,W；
calculating a weighted canonical decision matrix; setting a weight according to the importance degree of the factors influencing the travel experience, wherein the weight of arrival time is maximum, the weight of the number of passengers is minimum, and a weight vector Q= [ r ] is formed 1 ,r 2 ,...,r W ] T Further, a weighted canonical matrix c= (C) is formed xy ) L.W, wherein c xy ＝r x ·b xy (x=1, 2., L; y=1, 2., W; r represents a weight coefficient);
determining an ideal solution and a negative ideal solution; determining a positive ideal solution c y + (y-th property of positive ideal solution) and negative ideal solution c y - (y-th attribute of negative ideal solution) then:
ideal solutiony＝1,2...,W；
Negative ideal solutiony＝1,2...,W；
Wherein, the carriage crowding degree, the number of waiting vehicles and the arrival time are all cost property, the positive ideal solution is the minimum value, and the negative ideal solution is the maximum value;
calculating the distance from each travel scheme to an ideal solution and a negative ideal solution;
distance to the ideal solution isx＝1,2...L；
Distance to negative ideal solution isx＝1,2...L；
Calculating queuing index values of all schemes, namely comprehensive evaluation indexes: the comprehensive evaluation index queuing value Que of the scheme is given by the distance from the positive ideal solution and the negative ideal solution x The calculation method of (2) is as follows:
x＝1,2...,L；
according to Que x Good and bad order of big-to-small arrangement scheme: the finally obtained comprehensive evaluation index value of each scheme, namely Que x From small to large, where Que x The largest scheme is the optimal scheme, namely the final output of the algorithm.
Further, the recommended travel scheme is a series of recommended schemes from the best match to the common match degree.
A travel scheme recommendation system comprises an interaction module, a platform waiting number judging module, a crowdedness judging module, a arrival time judging module and a travel scheme judging module;
the interaction module is used for receiving query information of a user and pushing an optimal trip scheme to the user;
the platform waiting number judging module is used for acquiring the time sequences of the people on and off the same route, the same platform and the same time period and predicting the waiting number of the platform;
the crowding degree judging module is used for statistically predicting the crowding degree of the carriage by combining the number of waiting vehicles at the platform and the number of passengers at the carriage;
the arrival time judging module is used for predicting the arrival time of the vehicle;
the travel scheme judging module is used for integrating the attributes of the waiting number of the platform, the crowding degree of the carriage and the arrival time, establishing a standardized decision matrix, giving weight to each attribute, calculating a weighted standardized decision matrix, and giving recommendation of the optimal travel scheme after matching with the inquiry information of the user.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. according to the method, the degree of congestion of the carriage is creatively added as a reference factor for proposal recommendation on consideration of time, and the combination of the number of passengers getting on and off the platform and the information amount obtained by video in the middle of the carriage is provided as a congestion degree judgment standard, so that the real-time congestion degree of each bus is accurately obtained, the requirements of more passengers can be met, and the method is more humanized.
2. According to the invention, a time sequence model is established when the number of passengers waiting at the platform is predicted, the ARMA model is utilized to predict the number of passengers getting on and off the platform, and meanwhile, reference factors such as weather, holidays and the like are added, so that a prediction result is more accurate.
3. When the method predicts the degree of the crowding of the carriage, the degree of the crowding of the carriage is judged by combining the number of people getting on and off the carriage and the HOG image, and the two are mutually complemented, so that the accuracy of judging the degree of the crowding can be improved.
4. The travel scheme recommendation provided by the invention is a series of travel recommendations from preference to general, and the possibility of free selection of passengers is ensured while the demands of the passengers are fully considered.
Drawings
Fig. 1 is a schematic diagram of a travel plan recommendation system of the present invention.
Fig. 2 is a flow of judgment by the congestion degree judgment module according to the present invention.
Fig. 3 is a flowchart of the travel plan recommendation method of the present invention predicting arrival time.
Fig. 4 is a flowchart of a travel plan recommending method of the present invention.
Detailed Description
For a better understanding of the technical solution of the present invention, examples provided by the present invention are described in detail below with reference to the accompanying drawings, but embodiments of the present invention are not limited thereto.
Examples
When a user selects a destination of travel, the system sorts all schemes by using a TOPSIS method after obtaining the predicted congestion degree and arrival time of each scheme carriage and recommending the optimal scheme to the user according to the travel habit of the user, such as the comfort level that part of the users prefer to travel, namely, the lower congestion degree of the carriage is hoped, the weight of each attribute is updated in real time, the higher weight is given to the congestion degree of the carriage.
The embodiment discloses a bus trip scheme recommending method, and all steps of the embodiment are completed based on a JAVA development environment, an OpenCV function library and a MySQL database. The specific implementation technical scheme of the invention is that firstly, inquiry information of a user is input to an interaction module of a travel scheme recommendation system, the travel scheme recommendation system further comprises a platform waiting number judging module, a crowding degree judging module, a arrival time judging module and a travel scheme judging module, and when the user inputs riding information, the real-time riding system of the optimal travel scheme is comprehensively calculated according to the platform waiting number, the carriage crowding degree and the arrival time.
(1) The crowding degree judgment mainly adopts the feature extraction of a direction gradient histogram (Histogram of Oriented Gradient, HOG) and the information quantity calculation, and judges the crowding degree of the carriage by combining the statistics of the number of waiting vehicles at the platform and the number of passengers in the carriage. And extracting HOG characteristics of an input video image in the middle of the carriage, setting a certain threshold value for each image partition, screening out effective gradient values, counting an angle histogram of the effective gradient, dividing the image into partitions, taking g angle directions for each partition, and calculating gradient values of the g angles. Variance G of combined effective gradient values 2 And the image information entropy H is calculated as follows:
h represents an h-th angular direction of the g angular directions; mu is the average gradient value, P ij The comprehensive characteristic of gray value at a certain pixel position and gray distribution of pixels around the gray value is represented, the field gray average value of the image is selected as the space characteristic value of gray distribution, and a characteristic binary group is formed by the gray value of the pixels of the image and the gray value of the pixels of the image, and is marked as (i, j), wherein i representsThe gray value (i is more than or equal to 0 and less than or equal to 255) of the pixel, j represents the field gray value (j is more than or equal to 0 and less than or equal to 255), f (i, j) is the frequency of occurrence of the characteristic binary group (i, j), U, V is the length and the width of the image respectively, and the information entropy of the image is H.
And E= (G, H) calculating the confusion degree of the final image pixels, wherein when the variance of the gradient values of the subareas is small and the information entropy is large, the pixel value distribution of the subareas is relatively chaotic, and the congestion degree of the subareas is large. The statistics results of the number of waiting vehicles at the platform and the number of passengers in the carriage are mutually complemented with the judgment results of the crowding degree in the middle of the carriage, and the corresponding statistics values of the number of passengers are larger when the carriage is crowded, so that the final carriage crowding degree is cooperatively judged by the statistics of the number of passengers of cameras at the front and rear doors of the carriage and the calculation results of the crowding degree of the cameras in the middle of the carriage, and the final carriage crowding degree is used as the judgment basis of a travel scheme.
Specifically, the step of predicting the car congestion degree includes:
calculating the real-time crowding degree: the method mainly judges according to the accumulated value of the number of people getting on and off the bus after the bus is taken, and corrects the result by utilizing the crowding degree calculated by the camera in the middle of the bus, wherein the crowding degree of the bus is judged in advance in an auxiliary mode by extracting the image partition HOG characteristic through the camera in the middle of the bus, and the method is calculated as follows:
wherein E is real_time The chaotic degree of the image pixels is queried in real time, namely the real-time crowding degree;is the accumulated value of the number of people getting on and off after departure; area is the Area that can stand; e (E) DIP The result of the image confusion degree judged by the camera in the middle of the carriage;
calculating the congestion degree of the arrival station: from real-time congestion level E real_time The station between the station where the inquiring vehicle is located and the passenger station is accumulated, the history statistics result of the number of passengers getting on and off the intermediate station is predicted, and the prediction of the arrival congestion degree can be realized, and the calculation is as follows:
wherein E is getin_station Is the degree of congestion of the arrival at the station,the number of passengers getting on or off the bus from the searched station to the station where the passenger is located.
(2) In the bus arrival time prediction part, the embodiment obtains an arrival time prediction value based on a large amount of bus arrival history data through a nonparametric regression method, and the arrival time prediction value is used as an input parameter of a Kalman filtering algorithm to realize dynamic correction of the arrival time so as to accurately predict the bus arrival time.
The nonparametric regression method predicts the state at the next moment by searching a large amount of historical data to find the set most similar to the current input state. Let the bus arrive at k-1 station with the data state corresponding to the start time of S. By T km And T' km And respectively representing the inter-station travel time from k-1 station to k station and the stop time of k station corresponding to the m-th record with the data state of S in the historical database. Taking J as the neighbor number of the non-parametric regression, and D as the dimension of the non-parametric regression. Using Euclidean distance a m Matching the data with historical data, selecting J neighbors (the set of the J neighbors is denoted by A), and predicting the travel time T of the bus from the k station to the k+1 station by using a weighted average method according to the neighbors k :
Similarly, the stop time of the bus at the station k can be predicted by using the formulas through the stop time of the stations k-d, …, k-2 and k-1, and d is the station number different from the station k. The sum of the predicted stop time of the bus at the k-1 station and the predicted running time from the k-1 station to the k station is the predicted running time and is used as an input parameter of a Kalman filtering algorithm.
The Kalman filtering algorithm is to make an estimation meeting the minimum mean square error for the signal to be processed by the established state equation and the observation equation, complete the estimation of the state variable by combining the estimation value of the last moment and the observation value of the current moment, and solve the current estimation value. The state equation of the Kalman filter for bus arrival time prediction is as follows:
wherein t is (k+1) 、t' (k+1) The method comprises the steps of respectively representing the time and the observation time of a bus from a start station to a kth station, u (k) represents the inter-station travel time of the bus from k to k+1 stations, S (k), I (k) and M (k) are respectively data state transition variables, input variables and measured value coefficients, w (k) and v (k) are respectively input noise and measured noise, the coefficients of the state transition variables, the input variables and the measured value of Kalman filtering are all 1, and the input noise and the measured noise are independent white noise with the average value of 0.
When the bus reaches the kth station, kalman filtering is performed according to the observed value T of the kth station k m+1 And calculating the optimal estimated value of the kth site by the k-1 basic time sequences, further obtaining the adjustment value of the kth+1 site, sequentially updating the time of the subsequent site according to the adjustment value, and adding the updated time sequence into the basic time sequence.
According to the current collected historical data, the inter-station travel time of buses on the same line is quite regular and can be circulated, so that on the premise of a large amount of historical data, a set which is most similar to the current input state can be found, the state at the next moment is predicted, the arrival time is further predicted, the Kalman filtering method is used for correcting the most adjacent historical data, good instantaneity is achieved, prediction errors caused by the fact that relatively long-term data are used are avoided to a certain extent, and the two methods complement each other, so that more accurate arrival time and total travel time for passengers to arrive at a destination can be predicted.
(3) The number of passengers is predicted, and the acquired data is divided according to routes, platforms, number of weeks and time periods to obtain a time sequence of the number of passengers getting on or off the same platform in the same route and the same time period. And establishing an ARMA model, and further predicting the number of waiting vehicles. The modeling step is as follows:
(1) testing the stability of the sequence: the acquired data is used for generating a time sequence diagram, the stability of the data is judged by observing the time sequence diagram, and according to the definition of the stability, the stable sequence has the properties of constant mean and constant variance, so that the time sequence diagram of the stable sequence fluctuates around a constant value, and the fluctuation range is bounded. Meanwhile, according to actual conditions, the number of passengers getting on and off the platform in a short time is presumed not to obviously increase or decrease.
(2) Model identification and grading: the general ARMA (p, q) model is in the form of:
wherein the method comprises the steps ofθ 1 ，θ 2 ，…，θ q And epsilon t ，ε t-1 ，…，ε 1 Respectively an autoregressive parameter, a moving average parameter and a white noise sequence, X t-1 ...X t-p Is the input argument sequence of the model, i.e. the history basis of the prediction. The steps p and q of the ARMA model are primarily judged according to the autocorrelation coefficients and the partial autocorrelation coefficients, and the sample Autocorrelation Coefficients (ACF) and the Partial Autocorrelation Coefficients (PACF) are calculated.
(3) Parameter estimation of the model: and carrying out parameter estimation on the model by adopting a least square method according to the identified model and the order thereof.
(4) And (3) model inspection: after model identification, order determination and parameter estimation, residual sequence inspection analysis is required to be carried out on the model. The fitting model is only valid when the residual sequence is a white noise sequence, and can be used for prediction. Calculating the difference sequence { ε (t)) S-step autocorrelation coefficient ρ 1 、ρ 2 、...ρ s Then construct chi-square statistic F s F is shown in the following formula s Subject to degrees of freedom s beta 2 Distribution; wherein n is the capacity of the residual sequence;
(5) and predicting by using the ARMA model calculation result. And after inputting data such as lines, platforms, time and the like, the predicted waiting number can be obtained. And taking the ARMA model calculation result as a main prediction basis, and simultaneously adding reference factors such as weather, holidays and the like, so that the prediction result is more accurate.
(4) And the travel scheme is judged, and three indexes of the compartment crowding degree, the number of passengers waiting for the bus at the platform and the bus arrival time are comprehensively considered, so that a riding scheme is scientifically and reasonably formulated for passengers, the riding satisfaction degree is improved, and intelligent travel is realized.
TOPSIS is a ranking method that approximates ideal solutions by ranking each solution in a set of solutions with the help of ideal solutions and negative ideal solutions for a multi-objective decision problem. The method comprises the steps of quantifying three indexes of the compartment crowding degree, the waiting number and the arrival time of each bus, giving corresponding weights according to actual conditions, respectively calculating the distance from each bus to an ideal solution and a negative ideal solution, sequencing all the schemes, and giving the optimal solution. The method comprises the following specific steps:
(1) standardizing a decision matrix; let the travel scheme decision matrix be z= (Z xy ) L.W, wherein x is a travel plan subscript, y is a plan attribute subscript, z xy The values representing the y-attributes of the x-scheme, L, W represent the length and width of the decision matrix, respectively, i.e. the total number of schemes is the same as the total number of attributes of each scheme. The scheme attributes comprise the contents of compartment crowdedness, waiting number, arrival time, transfer times and the like. Let normalized decision matrix b= (B) xy ) L.W, where b xy Values representing the normalized x schema y attribute are:
x＝1,2...,L；y＝1,2...,W；
(2) calculating a weighted canonical decision matrix; setting a weight according to the importance degree of the factors influencing the travel experience, wherein the weight of arrival time is maximum, the weight of the number of passengers is minimum, and a weight vector Q= [ r ] is formed 1 ,r 2 ,...,r W ] T Further, a weighted canonical matrix c= (C) is formed xy ) L.W, wherein c xy ＝r x ·b xy X=1, 2., L; y=1, 2., W; r represents a weight coefficient;
(3) determining an ideal solution and a negative ideal solution; determining a positive ideal solution c y + (y-th property of positive ideal solution) and negative ideal solution c y - (y-th attribute of negative ideal solution) then:
ideal solutiony＝1,2...,W；
Negative ideal solutiony＝1,2...,W；
The method comprises the steps of determining the degree of carriage crowding, waiting number and arrival time, wherein the degree of carriage crowding, waiting number and arrival time are all cost properties, the positive ideal solution is the minimum value, and the negative ideal solution is the maximum value.
(4) Calculating the distance from each travel scheme to an ideal solution and a negative ideal solution;
distance to the ideal solution isx＝1,2...L；
Distance to negative ideal solution isx＝1,2...L；
(5) Calculating queuing index values (i.e., comprehensive evaluation indexes) of each scheme: given together by the distances to the positive ideal solution and the negative ideal solution, the calculation method is as follows:
x＝1,2...,L；
(6) according to Que x Good and bad order of big-to-small arrangement scheme: the finally obtained comprehensive evaluation index value of each scheme, namely Que x From small to large, where Que x The largest scheme is the optimal scheme, namely the final output of the algorithm.
The transfer of buses can be analogized to subways, so the method of the embodiment can also be applied to subways.
The above examples are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above examples, and any other changes, modifications, substitutions, combinations, and simplifications that do not depart from the spirit and principle of the present invention should be made in the equivalent manner, and the embodiments are included in the protection scope of the present invention.
Claims (9)
1. The travel scheme recommending method is characterized by comprising the following steps of:
receiving inquiry information of a user;
predicting the number of waiting persons at a platform: acquiring time sequences of the number of passengers getting on and off the same route, the same platform and the same time period, and predicting the number of passengers waiting at the platform;
the step of predicting the number of bus stops is as follows:
collecting the number of people getting on or off the bus through cameras of front and rear doors of the bus, and extracting the number of people getting on or off the bus at the same platform, the same line and the same time period after the data volume reaches a certain degree to form a number sequence;
after checking the stability, a time sequence model is established, the ARMA model is utilized to predict the number of passengers on and off the platform, and meanwhile, weather and holiday reference factors are added;
predicting the degree of congestion of the carriage: the number of waiting vehicles at the platform and the number of passengers in the carriage are combined for statistics and prediction of the degree of carriage crowding;
predicting the arrival time of a vehicle;
and (3) judging an optimal scheme: and (3) integrating the attributes of the number of waiting vehicles at the platform, the degree of congestion of the carriage and the arrival time, establishing a standardized decision matrix, assigning weights to the attributes, calculating a weighted standardized decision matrix, and giving out the recommendation of the optimal travel scheme after matching the user inquiry information.
2. The travel plan recommendation method according to claim 1, wherein the modeling step of the ARMA model is as follows:
stability of the population sequence: generating a time sequence diagram by using the acquired data, judging the stability of the data by observing the time sequence diagram, wherein the stable sequence has the properties of constant mean and constant variance according to the definition of the stability, so that the time sequence diagram of the stable sequence fluctuates around a constant value, and the fluctuation range is bounded; meanwhile, according to actual conditions, the number of passengers getting on or off the platform is presumed not to be obviously increased or decreased in a short time;
model identification and grading: the ARMA (p, q) model is of the form:
wherein phi is 1 ，φ 2 ，…，φ p 、θ 1 ，θ 2 ，…，θ q And epsilon t ，ε t-1 ，…，ε 1 Respectively an autoregressive parameter, a moving average parameter and a white noise sequence, X t-1 ...X t-p The input independent variable sequence of the model is the history basis of prediction; preliminary judging the orders p and q of the ARMA model according to the autocorrelation coefficients and the partial autocorrelation coefficients, and calculating the sample autocorrelation coefficients and the partial autocorrelation coefficients;
parameter estimation of the model: according to the identified model and the order thereof, carrying out parameter estimation on the model by adopting a least square method;
and (3) model inspection: performing model identification, order determination and parameter estimation, and performing residual sequence inspection analysis on the model; pseudo when the residual sequence is a white noise sequenceThe composite model is valid and can be used for prediction; s-step autocorrelation coefficient ρ of residual sequence { ε (t) } is calculated 1 、ρ 2 、...ρ s Then construct chi-square statistic F s F is shown in the following formula s X subject to a degree of freedom s 2 Distribution; wherein n is the capacity of the residual sequence;
predicting by using ARMA model calculation results; after inputting the line, platform and time data, the predicted waiting number can be obtained; taking the ARMA model calculation result as a main prediction basis, and simultaneously adding weather and holiday reference factors.
3. The travel plan recommendation method according to claim 1, wherein the step of predicting the degree of congestion of the vehicle compartment is:
extracting image partition HOG characteristics through a camera in the middle of the carriage, and pre-judging the crowding degree of the carriage;
capturing heads of people through a carriage door camera, counting the number of people getting on and off, and obtaining the number of people in the carriage by accumulating the number of people getting on and off each platform and subtracting the number of people getting off;
and calculating the arrival crowdedness degree by combining the pre-judging result and the number of the carriages.
4. The travel plan recommendation method according to claim 3, wherein the extracting image partition HOG features, and the pre-judging the congestion degree of the carriage specifically comprises:
adopting a direction gradient histogram, extracting features, calculating information quantity, extracting HOG features of an input video image in the middle of a carriage, setting a certain threshold value for each image partition, screening out an effective gradient value, counting an angle histogram of an effective gradient, dividing the image into partitions, taking g angle directions for each partition, and calculating gradient values of the g angles; variance G of combined effective gradient values 2 And image information entropy H calculationThe following are provided:
h in the above formula represents the h-th angle direction in the g angle directions; mu is the average gradient value; p (P) ij A comprehensive feature representing gray values at a certain pixel position and gray distribution of pixels around the gray values; selecting the field gray average value of the image as a space characteristic value of gray distribution, and forming a characteristic binary group with the pixel gray of the image, wherein i represents the gray value of the pixel, i is more than or equal to 0 and less than or equal to 255, j represents the field gray value, and j is more than or equal to 0 and less than or equal to 255; f (i, j) is the frequency at which feature doublet (i, j) occurs; u, V is the length and width of the image, respectively; the information entropy of the image is H;
calculating the confusion degree E= (G, H) of the final image pixels, and when the partition gradient value variance is small and the information entropy is large, indicating that the pixel value distribution of the partition is more chaotic, then the partition congestion degree is large; the statistics results of the number of waiting vehicles at the platform and the number of passengers in the carriage are mutually complemented with the judgment results of the crowding degree in the middle of the carriage, and the corresponding statistics values of the number of passengers are larger when the carriage is crowded, so that the final carriage crowding degree is cooperatively judged by the statistics of the number of passengers of cameras at the front and rear doors of the carriage and the calculation results of the crowding degree of the cameras in the middle of the carriage, and the final carriage crowding degree is used as the judgment basis of a travel scheme.
5. The travel plan recommendation method according to claim 1, wherein the step of predicting the arrival time of the vehicle is:
calculating Euclidean distance between current data and historical data under the condition of original data matching by adopting a historical state matching method, so as to predict arrival time;
the nonparametric regression method searches a large amount of historical data to find the set most similar to the current input state so as toPredicting the state of the next moment; assuming that the bus arrives at a k-1 station, and the data state corresponding to the starting time is S; by T km And T' km Respectively representing the inter-station travel time from k-1 station to k station and the stop time of k station corresponding to the m-th record with the data state of S in the historical database; taking J as the neighbor number of the non-parametric regression, and D as the dimension of the non-parametric regression; using Euclidean distance a m Matching the bus with historical data, selecting J neighbors, wherein A represents a set of the J neighbors, and predicting the travel time T of the bus from k stations to k+1 stations according to the neighbors by using a weighted average method k ：
e is a neighbor in a set of a different from m;
similarly, the stop time of the bus at the station k can be predicted by using the formulas through the stop time of the stations k-d, …, k-2 and k-1, and d is the station number different from the station k.
6. The travel plan recommending method according to claim 5, wherein when the arrival time of the vehicle is predicted, the sum of the predicted stop time of the bus at the current station and the predicted travel time to the next station is taken as the predicted travel time, and the sum is taken as an input parameter of a kalman filter algorithm, so that the dynamic correction of a predicted result is realized;
the sum of the predicted stop time of the bus at the k-1 station and the predicted running time from the k-1 station to the k station is the predicted running time and is used as an input parameter of a Kalman filtering algorithm;
the Kalman filtering algorithm is to make an estimation meeting the minimum mean square error for the signal to be processed by the established state equation and the observation equation, complete the estimation of the state variable by combining the estimation value of the last moment and the observation value of the current moment, and solve the current estimation value; the state equation of the Kalman filter for bus arrival time prediction is as follows:
wherein t is (k+1) 、t' (k+1) Respectively representing the time and the observation time of the bus from the start station to the (k+1) th station, u (k) represents the inter-station travel time of the bus from k to the (k+1) th station, S (k+1) 、I (k+1) And M (k+1) Respectively, data state transition variable, input variable, measured value coefficient, w (k+1) And v (k+1) The method comprises the steps of respectively inputting noise and measuring noise, setting the state transition variable, the input variable and the coefficient of a measured value of Kalman filtering to be 1, wherein the input noise and the measuring noise are independent white noise which is uncorrelated and has the average value of 0;
when the bus reaches the kth station, kalman filtering is performed according to the observed value T of the kth station k m+1 And calculating the optimal estimated value of the kth site by the k-1 basic time sequences, further obtaining the adjustment value of the kth+1 site, sequentially updating the time of the subsequent site according to the adjustment value, and adding the updated time sequence into the basic time sequence.
7. The travel plan recommendation method according to claim 1, wherein the step of determining the optimal plan is:
standardizing a decision matrix; let the travel scheme decision matrix be z= (Z xy ) L·W Wherein x is a travel scheme index, y is a scheme attribute index, z xy Values representing the y attributes of the x schemes, L, W represent the length and width of the decision matrix, i.e. the total number of schemes and the total number of attributes of each scheme, respectively; the scheme attributes comprise the degree of carriage crowding, the number of passengers waiting for the carriage, the arrival time and the transfer times; let normalized decision matrix b= (B) xy ) L·W Wherein b xy Values representing the normalized x schema y attribute are:
calculating a weighted canonical decision matrix; setting a weight according to the importance degree of the factors influencing the travel experience, wherein the weight of arrival time is maximum, the weight of the number of passengers is minimum, and a weight vector Q= [ r ] is formed 1 ,r 2 ,…,r L ] T Further, a weighted canonical matrix c= (C) is formed xy ) L·W Wherein c xy ＝r x ·b xy ，x＝1,2...,L；y＝1,2...,W；r x Representing the weight coefficient;
determining an ideal solution and a negative ideal solution; determining a positive ideal solution c y + And negative ideal solution c y - The method comprises the steps of carrying out a first treatment on the surface of the Then:
ideal solution c y +
Negative ideal solution c y -
Wherein, the carriage crowding degree, the number of waiting vehicles and the arrival time are all cost property, the positive ideal solution is the minimum value, and the negative ideal solution is the maximum value;
calculating the distance from each travel scheme to an ideal solution and a negative ideal solution;
distance to the ideal solution is
Distance to negative ideal solution is
Calculating queuing index values of all schemes, namely comprehensive evaluation indexes: the comprehensive evaluation index queuing value Que of the scheme is given by the distance from the positive ideal solution and the negative ideal solution x The calculation method of (2) is as follows:
according to Que x Good and bad order of big-to-small arrangement scheme: the finally obtained comprehensive evaluation index value of each scheme, namely Que x From small to large, where Que x The largest scheme is the optimal scheme, namely the final output of the algorithm.
8. The travel plan recommending method according to claim 7, wherein the recommended travel plan is a series of recommended plans from best matching to matching degree general.
9. The travel scheme recommendation system is characterized by comprising an interaction module, a platform waiting number judging module, a crowdedness judging module, a arrival time judging module and a travel scheme judging module;
the interaction module is used for receiving query information of a user and pushing an optimal trip scheme to the user;
the platform waiting number judging module is used for acquiring the time sequences of the people on and off the same route, the same platform and the same time period and predicting the waiting number of the platform;
the step of predicting the number of bus stops is as follows:
collecting the number of people getting on or off the bus through cameras of front and rear doors of the bus, and extracting the number of people getting on or off the bus at the same platform, the same line and the same time period after the data volume reaches a certain degree to form a number sequence;
after checking the stability, a time sequence model is established, the ARMA model is utilized to predict the number of passengers on and off the platform, and meanwhile, weather and holiday reference factors are added;
the crowding degree judging module is used for statistically predicting the crowding degree of the carriage by combining the number of waiting vehicles at the platform and the number of passengers at the carriage;
the arrival time judging module is used for predicting the arrival time of the vehicle;
the travel scheme judging module is used for integrating the attributes of the waiting number of the platform, the crowding degree of the carriage and the arrival time, establishing a standardized decision matrix, giving weight to each attribute, calculating a weighted standardized decision matrix, and giving recommendation of the optimal travel scheme after matching with the inquiry information of the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910378683.7A CN110245377B (en) | 2019-05-08 | 2019-05-08 | Travel scheme recommendation method and recommendation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910378683.7A CN110245377B (en) | 2019-05-08 | 2019-05-08 | Travel scheme recommendation method and recommendation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110245377A CN110245377A (en) | 2019-09-17 |
| CN110245377B true CN110245377B (en) | 2023-08-04 |
Family
ID=67883816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910378683.7A Active CN110245377B (en) | 2019-05-08 | 2019-05-08 | Travel scheme recommendation method and recommendation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110245377B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110782694B (en) * | 2019-10-31 | 2021-10-08 | 同济大学 | Internet bus arrival and stop management optimization method |
| CN111091231B (en) * | 2019-11-25 | 2022-04-15 | 珠海格力电器股份有限公司 | Prediction model training method, time prediction method, training device and terminal |
| CN111415679B (en) * | 2020-03-25 | 2023-02-28 | Oppo广东移动通信有限公司 | Site identification method, device, terminal and storage medium |
| CN112256967A (en) * | 2020-10-27 | 2021-01-22 | 北京字跳网络技术有限公司 | Travel recommendation method, device, equipment and storage medium |
| CN114299746B (en) * | 2021-12-30 | 2023-04-14 | 武汉长飞智慧网络技术有限公司 | Image recognition-based bus scheduling method, device and medium |
| CN115588298B (en) * | 2022-10-28 | 2023-12-29 | 广州地铁集团有限公司 | Urban rail passenger flow broadcasting induction method based on machine vision |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014010493A (en) * | 2012-06-27 | 2014-01-20 | Koito Electric Industries Ltd | Bus stop information service system |
| CN105096584A (en) * | 2014-05-06 | 2015-11-25 | 中兴通讯股份有限公司 | Traffic decision support method, device, and system |
| CN105702035A (en) * | 2016-03-23 | 2016-06-22 | 武汉大学 | Method for evaluating bus taking difficulty degree by using historical bus data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140012498A1 (en) * | 2012-03-19 | 2014-01-09 | Thomas F. Gustafson | Informed Traveler Program and Application |
-
2019
- 2019-05-08 CN CN201910378683.7A patent/CN110245377B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014010493A (en) * | 2012-06-27 | 2014-01-20 | Koito Electric Industries Ltd | Bus stop information service system |
| CN105096584A (en) * | 2014-05-06 | 2015-11-25 | 中兴通讯股份有限公司 | Traffic decision support method, device, and system |
| CN105702035A (en) * | 2016-03-23 | 2016-06-22 | 武汉大学 | Method for evaluating bus taking difficulty degree by using historical bus data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110245377A (en) | 2019-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245377B (en) | Travel scheme recommendation method and recommendation system | |
| CN108286980B (en) | Method for predicting destination and recommending driving route | |
| CN107945507B (en) | Travel time prediction method and device | |
| CN112466122B (en) | Method and device for generating alternative line set and planning line of public traffic line network | |
| CN107545320B (en) | Urban rail transit passenger path planning method and system based on graph theory | |
| CN105677804B (en) | Method and device for determining authoritative site and establishing database of authoritative site | |
| CN115409388B (en) | Multi-vehicle type customized bus operation optimization method | |
| WO2019042297A1 (en) | Method and system for planning optimal route and optimal driving manner of electric automobile | |
| CN108009972B (en) | Multi-mode travel O-D demand estimation method based on multi-source data check | |
| CN110836675B (en) | Decision tree-based automatic driving search decision method | |
| CN111932084A (en) | System for assessing accessibility of urban public transport | |
| CN110348614B (en) | Method for obtaining passenger OD and method for predicting bus passenger flow | |
| CN108038578B (en) | Public bicycle static scheduling method based on demand prediction and central radiation network | |
| CN112556717B (en) | Travel mode screening method and travel route recommending method and device | |
| Yamamoto et al. | Structured random walk parameter for heterogeneity in trip distance on modeling pedestrian route choice behavior at downtown area | |
| CN111582527A (en) | Travel time estimation method and device, electronic equipment and storage medium | |
| CN113408833A (en) | Public traffic key area identification method and device and electronic equipment | |
| CN111723871B (en) | Estimation method for real-time carriage full load rate of bus | |
| CN112351394A (en) | Traffic travel model construction method based on mobile phone signaling data | |
| CN114692015A (en) | Riding point recommendation method based on density clustering | |
| CN114724414B (en) | Method and device for determining urban air traffic sharing rate, electronic equipment and medium | |
| CN113592246B (en) | PCA-TOPSIS model-based road construction period bus network evaluation method | |
| CN113962434A (en) | Dynamic large-scale ride-sharing route planning method based on robust path | |
| CN115357798A (en) | Big data-based smart travel target recommendation method and system | |
| CN112766950A (en) | Dynamic path cost determination method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |