CN106022541B - Arrival time prediction method - Google Patents
Arrival time prediction method Download PDFInfo
- Publication number
- CN106022541B CN106022541B CN201610472352.6A CN201610472352A CN106022541B CN 106022541 B CN106022541 B CN 106022541B CN 201610472352 A CN201610472352 A CN 201610472352A CN 106022541 B CN106022541 B CN 106022541B
- Authority
- CN
- China
- Prior art keywords
- neural network
- station
- models
- time
- data
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000013528 artificial neural network Methods 0.000 claims abstract description 21
- 238000003062 neural network model Methods 0.000 claims description 85
- 238000012549 training Methods 0.000 claims description 52
- 230000001537 neural effect Effects 0.000 claims description 38
- 238000012360 testing method Methods 0.000 claims description 30
- 210000002569 neuron Anatomy 0.000 claims description 26
- 238000001914 filtration Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 2
- 238000013530 stochastic neural network Methods 0.000 abstract description 20
- 230000001413 cellular effect Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 6
- 238000013459 approach Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- 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
-
- 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
- G08G1/127—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 to a central station ; Indicators in a central station
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
- Navigation (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
A system and a method for predicting arrival time based on a random neural network group are disclosed, wherein a vehicle-mounted action device periodically acquires positioning information (longitude and latitude coordinates), compares the positioning information with site information (a polygonal or circular area) in the vehicle-mounted action device to judge whether the vehicle-mounted action device arrives at the site (or leaves the site), and reports the arrival (or leaving) information and a time point to a cloud server. The cloud server is responsible for collecting arrival (or departure) information returned by the vehicle-mounted mobile device, analyzing travel time among all the stations, storing the data into the cloud database server, and using a travel time data set to train parameter values of the stochastic neural network group algorithm.
Description
The invention is a divisional application of an invention application with the application date of 2015, 3/3, application number of 201510094269.5 and the name of 'a station arrival time prediction system and method'.
Technical Field
The invention relates to the technical field of wireless communication, in particular to a method for predicting arrival time.
Background
At present, the prior art for predicting the arrival time of public transport means is as follows: historical data is used for statistics and averaging, and the average vehicle speed and the travel time between stations are obtained, or the current vehicle real-time instantaneous vehicle speed information is applied to estimate the station arrival time. However, these methods cannot reflect the real-time traffic status variation between the stations, and thus cause a large arrival time information error.
Taiwan patent publication No. TW201137803 mainly proposes to collect arrival information reported by buses in the past to estimate average vehicle speed and travel time between arrival and arrival, and to perform statistics according to different weeks and time periods, so as to obtain historical average vehicle speed and travel time when a user inquires. Although the method can quickly provide the estimated arrival time, the average value of the historical data is mainly adopted, so that the arrival time cannot be predicted according to the real-time road conditions, and a large error may be caused in the prediction of the arrival time.
The taiwan patent publication No. TW201344647 predicts the arrival time, and provides a vehicle speed adjustment suggestion to the driver according to the real-time location information of the bus, thereby improving the arrival accuracy. Although the method can estimate the arrival time and also provide the arrival point control, the method mainly adopts the average value of historical data, and the arrival time cannot be predicted according to the real-time road condition, so that a large error in the prediction of the arrival time can be caused.
Taiwan patent publication No. TW201405497 mainly proposes that when a vehicle-mounted mobile device passes through each road segment, the travel time of each road segment is reported back to a monitoring center at the back end in real time through the vehicle-mounted mobile device, and then the monitoring center distributes the shortest travel time and the longest travel time of each road segment to all vehicle-mounted mobile devices. And if the travel time of the vehicle-mounted mobile equipment is between the shortest travel time and the longest travel time, no report is given. Although the method can effectively master the travel time of each road section and reduce the transmission quantity, a prediction method of the arrival time of the bus is not provided, so that the arrival information of the bus cannot be predicted.
Taiwan patent publication No. TW201117146 mainly provides a method for inquiring the travel time of a bus, so that a user can inquire the real-time location and the travel time of the bus he wants to take. Although the method can enable a user to inquire the real-time position and the travel time of the bus, a prediction method of the bus arrival time is not provided, so that the bus arrival information cannot be predicted.
Taiwan patent publication No. TW200828190 mainly proposes to use a mobile device of a user to receive station information, and to send a notification to remind the user when the user arrives at the station. Although this method can alert the user when the user arrives at the site, it can provide real-time arrival information, but it cannot provide predictive information.
Taiwan patent No. TWI252441 mainly proposes that a bus receives a satellite positioning signal, and transmits position information back to a monitoring center in real time, and a prediction module of the monitoring center predicts arrival time according to the real-time position of the bus. Although this method may provide arrival time prediction, reference to empirical values is only mentioned in the patent, and in the future, explicit reference is made to a method of predicting bus arrival times.
Taiwan patent with publication number TWI341998 mainly proposes to predict travel time according to the real-time speed of a bus and the distance from the bus to each station; and calculating walking time according to the walking speed of the user and the distance from the user to each station. And finally estimating a suitable station according to the travel time and the walking time. Although the method can provide a prediction method of the bus travel time, the method mainly considers the current real-time speed and the arrival distance of the bus, but the traffic information between the bus and the station is not considered, so that a large error in the prediction of the arrival time can be caused.
Taiwan patent publication No. TW201232489 proposes that a driving speed is predicted by using an empirical mode decomposition method of hilbert-yellow transform (HHT) in combination with a gray mode, and then converted into a travel time and an arrival time from the predicted driving speed. Although the method can effectively use the mathematical and statistical models to predict the vehicle speed, the method cannot avoid the influence of extreme values because all data are used for analysis, and large errors in the prediction of the arrival time can be caused.
Disclosure of Invention
In view of the above problems of the prior art, an object of the present invention is to provide a method for predicting arrival time, which collects travel time between stations of each link and each time period, and provides a novel random neural network group to analyze the travel time data set, establish a plurality of neural network models to avoid the influence of extreme values, and comprehensively consider the prediction results of the neural network models to improve the accuracy of prediction, so as to predict the arrival time of a bus to be taken by a user, and provide the prediction results to the user as a reference.
The arrival time prediction system comprises a plurality of station boards, a plurality of vehicle-mounted terminal devices, a plurality of cell network base stations, a cloud computing server, a cloud historical database and a plurality of arrival time prediction system client devices. Wherein, each station stop board has longitude and latitude coordinate information. When each vehicle-mounted terminal device approaches the plurality of station stop boards, each vehicle-mounted terminal device senses the plurality of longitude and latitude coordinate information and generates arrival information. The arrival information is transmitted through the cellular network base stations, the cloud computing server calculates the travel time after receiving the arrival information from the cellular network base stations, predicts the remaining travel time according to the travel time and the inquiry station, converts the remaining travel time into the arrival time, and transmits the arrival time through the cellular network base stations. The cloud historical database stores longitude and latitude coordinate information and travel time between station boards. The arrival time prediction system client equipment sends a query station, receives arrival time transmitted by a cell network base station and displays the arrival time.
The arrival time prediction method of the invention comprises the following steps: setting a random neural network group algorithm parameter value; reading station-to-station travel times in a historical database; randomly generating m neural network models; filtering out the neural network models with the accuracy lower than a threshold value, and then remaining k neural network models; acquiring real-time station-to-station travel time or test data in a test stage; inputting travel time or test data into the k filtered neural network models, and predicting station-to-station travel time; and converting the predicted station-to-station travel time into the arrival time of the target station.
In summary, the arrival time prediction system and method of the present invention have one or more of the following advantages:
1. the invention collects real-time station-to-station travel times for various road segments and time periods to estimate the travel time for the current vehicle location to reach the target station.
2. The invention provides a novel random neural network group to analyze the travel time data set, establish a plurality of neural network models, and comprehensively consider the prediction results of the neural network models to improve the prediction accuracy, so as to predict the arrival time of the bus to be taken by the user, and provide the prediction results for the user as reference.
3. In the learning stage of the random neural network group algorithm, a plurality of data are respectively taken out from the data set for each neural network model at random as training data, the rest data are taken as test data in the training stage, and the training data are input into each neural network model for learning, so that the influence of extreme values can be avoided.
4. In the testing stage and the implementation stage of the stochastic neural network group algorithm, the travel time predicted by each neural network model and the weight learned in the training stage are used for weighted average, finally, the travel time after weighted average is used as the travel time predicted value of the stochastic neural network group algorithm, and the travel time is converted into the arrival time, so that the arrival time prediction is carried out.
Drawings
FIG. 1 is a schematic diagram of a station arrival time prediction system according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a second arrival time prediction method according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a third arrival time prediction method according to an embodiment of the present invention;
FIG. 4 is a diagram of four types of neural network models according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a fifth embodiment of the invention for predicting travel time.
Detailed Description
Referring to fig. 1, the present invention relates to a system for arrival time prediction based on stochastic neural network clustering. The system can mainly predict the arrival time of the vehicle, is suitable for passenger transport operators, logistics operators or other related operators with the arrival time prediction requirement, provides the predicted arrival time for client equipment, and enables a client or a user to master vehicle information and arrival information in real time, thereby saving waiting time, and mainly comprises the following six modules: (1) multiple station stop boards 100: the station board device mainly comprises a group of longitude and latitude coordinate information, the information can be stored in the vehicle-mounted terminal device and the cloud computing server in advance, and when the vehicle-mounted terminal device approaches the station board, the vehicle-mounted terminal device can sense the station information. In addition, the stop board device can also be embedded with an RFID (Radio Frequency IDentification) tag, so that the stop board can be sensed when a vehicle approaches, and the arrival can be judged according to the detection. (2) The plurality of in-vehicle terminal apparatuses 101: the device mainly comprises a Global Positioning System (GPS) module, a cellular network module, and a database module (not shown in fig. 1), and can collect the current position of a vehicle (including longitude and latitude coordinates), and judge whether the current position is close to a station stop board 100, judge that a station is arrived if the current position is in a range near the station stop board 100, and transmit the arrival information and time point back to a cloud computing server terminal 103 through a cellular network base station 102. In addition, in the arrival judging section, the in-vehicle terminal device 101 may also be embedded with an RFID reader, may sense a stop board when the vehicle approaches, and may receive an RFID tag signal from the stop board device to judge whether the vehicle arrives. (3) A plurality of cellular network base stations 102: each cellular network base station 102 provides a data transmission function and a data reception function, and is responsible for data transmission between the vehicle-mounted terminal apparatus 101, the cloud computing server 103, and the arrival time prediction system client apparatus 106. (4) Cloud computing server 103: the server mainly can collect and analyze arrival information and arrival time points from the vehicle-mounted terminal device 101, calculate the travel time between each station and each station according to each arrival time point, input a data set of the travel time between a plurality of previous stations on the traveling route of the target station inquired by a user into the neural network group trained by the random neural network group arrival time prediction method provided by the invention, analyze and calculate to obtain the residual travel time prediction of the arrival of the target station, and convert the residual travel time prediction into the arrival time of the arrival of the target station. (5) Cloud history database 105: the database can mainly store historical travel time between each station and each station, and can be used as a training data set of a random neural network group to train each neural network model. (6) Multiple arrival time prediction system client devices 106: the equipment can be mobile equipment, is provided with a man-machine interaction interface and a network transmission module, and can enable a user to inquire and display the arrival time prediction of a target station to be acquired through the equipment. The station and time to be taken by the user can be preset by the user, and then the device actively updates and judges the station and time to be taken by the user, and actively sends out reminding information and sound to the user when the vehicle is about to arrive.
Referring to fig. 2 and 3, the present invention further provides a method for predicting arrival time based on a stochastic neural network. The method mainly comprises 2 stages: (a) a training phase and (b) a performance and testing phase. Wherein, the training phase mainly includes 4 steps, do respectively:
step S201: setting a random neural network group algorithm parameter value;
step S202: reading the travel time between each station in the historical database;
step S203: randomly generating m neural network models;
step S208: and after filtering out the neural network models with the accuracy lower than the threshold value, remaining k neural network models.
The implementation and test phases mainly comprise 3 steps, which are respectively:
step S301: acquiring real-time travel time between each station and each station or test data in a test stage;
step S302: inputting data into the filtered k neural network models, and predicting station-to-station travel time;
step S306: and obtaining the predicted station-to-station travel time, and converting the station-to-station travel time into the arrival time of the target station.
In step S201, relevant parameter values of the random neural network group algorithm are first set by the arrival time prediction system developer, where the relevant parameter values include the number of neural network models (m will be described later as an example), the maximum number of hidden layers in the neural network models (hmax will be described later as an example), the maximum number of neurons per hidden layer in the neural network models (cmax will be described later as an example), the proportion of the number of training data of the training neural network models to the total number of training stage data (r% will be described later as an example), and a correctness threshold (wthreshold will be described later as an example).
In step S202, the time when the vehicle arrives at each station is obtained from the cloud history database 103 and converted into the station-to-station travel time, for example: the arrival time of the station 1 is the time point t1, and the arrival time of the station 2 is the time point t2, then the travel time of the station 1 to the station 2 is | t2-t1 |. And then the travel time set is used as input and output data of the neural network model for subsequent learning. Taking fig. 1 as an example, to predict the time to reach station n when the vehicle travels to station n-2 (i.e., the travel time of the target output is | tn-tn-2|), the input travel time data set may include { | t2-t1|, | t3-t2|, · and | tn-2-tn-3| }.
In step S202, m neural network-like models are randomly generated according to the random neural network group algorithm parameter values set by the arrival time prediction system developer, and each neural network-like model uses r% of the total training data number randomly obtained as training and learning data, and uses the remaining data (i.e., 100% -r% of the data amount) as verification of each neural network-like model, and each neural network-like model trains and verifies the obtained different data. In addition, each neural network model will generate 0-hmax hidden layers and 0-cmax neurons for each hidden layer according to the parameter settings, where the combination of hidden layers and neurons for each neural network model will be different. Inputting the r% data into the neural network model for training and learning, after convergence is reached, inputting 100% -r% data (namely test data in the training stage) into the trained neural network model, obtaining the predicted travel time, and comparing the predicted travel time with the correct travel time, so as to obtain the correct rate of each neural network model, and taking the correct rate as the weight value in the implementation stage and the test stage.
In step S208, after the neural network models with the accuracy lower than the threshold are filtered, k neural network models remain: comparing the accuracy of m neural network models generated randomly with an accuracy threshold value wthreshold, and removing the neural network models (with low accuracy) lower than the threshold value to obtain k neural network models; if the accuracy of any neural network model is not higher than the threshold value, the process returns to step S201, the arrival time prediction system developer resets the threshold value, and the random neural network group is retrained.
In step S301, in the execution and test phase, first, the real-time station-to-station travel time of the vehicle acquired in advance is: when the vehicle moves to station n-2 in fig. 1, the user wants to inquire the arrival time prediction of station n (i.e., the travel time of the target output is | t |)n-tn-2|). At this time, the travel time data of the vehicle in the journey can be calculated to be set { | t2-t1|,|t3-t2|,...,|tn-2-tn-3And | as input data of the neural network-like model.
In step S302, the acquired real-time travel time data set { | t2-t1|,|t3-t2|,...,|tn-2-tn-3I is input into k neural network models after filtering, and each neural network model predicts one | t |)n-tn-2The predicted travel time | is multiplied by the weight values of the neural network models obtained in the training stage (i.e., the accuracy of the neural network models in the training stage), and the sum of the weighted values is divided by the sum of the weight values (i.e., weighted average).
In step S301, a predicted travel time | t obtained by comprehensively considering k neural network models is acquiredn-tn-2After | the vehicle is real-time at the time tn-2Plus predicted travel time | tn-tn-2And | obtaining the arrival time prediction of the arrival station n and providing the prediction result for the user.
The invention collects and analyzes the station arrival (departure) information (including station information, time point and the like) returned from the vehicle-mounted terminal device 101, converts the data set into the station arrival travel time, stores the station arrival information in the cloud historical database 105, designs and implements a station arrival information prediction method module based on the stochastic neural network group algorithm in the cloud computing server 103, can access the travel time set in the cloud historical database 105, inputs the travel time set into the station arrival information prediction method module based on the stochastic neural network group algorithm, and carries out neural network model training to predict the travel time. When the arrival time prediction system client performs station arrival time prediction, the previous station information of the current vehicle-mounted terminal device 101 in the route return can be input into the trained neural network group to perform travel time prediction of the arrival target station, and then the arrival time of the arrival target station is converted and provided to the arrival time prediction system client device 106. The technical feature of the present invention is to propose and design a stochastic neural network group algorithm, and apply it to the arrival information prediction method, which will be described below by way of example.
The invention provides a system for predicting arrival time based on a stochastic neural network group, and the system architecture is shown in figure 1. The system comprises a plurality of station boards 100, a plurality of vehicle-mounted terminal devices 101, a plurality of cellular network base stations 102, a cloud computing server 103, a cloud history database 105 and a plurality of arrival time prediction system client devices 106. In the present embodiment, a station stop board 100 of the same route is taken as an example, and there are n stations in the route, and each station has location information (including longitude and latitude). As shown in table one, the route 1 includes a total of 12 stations (i.e., n in fig. 1 is 12), and the corresponding longitude and latitude can be stored in the vehicle-mounted terminal device; when the vehicle number 1 travels from the station 1 to the station 2, the GPS module of the vehicle-mounted terminal device detects that the longitude of the vehicle is 120.97839 and the latitude is 24.808658 at 2014/4/114: 53, evaluates that the vehicle is close to the station 2 (for example, within 30 meters of the straight-line distance), determines that the vehicle is an arrival station, and transmits the arrival station information (including the station number and the time point) back to the cloud computing server 103 through the cellular network base station 102.
In addition, the station stop board 100 may also have an RFID tag, and the vehicle-mounted terminal device 101 may have an RFID reader, so that when the vehicle-mounted terminal device 101 approaches the station stop board 100, the RFID tag of the station stop board 100 can be detected, and the station is determined to arrive, and the arrival information (including the station number and the time point) is transmitted back to the cloud computing server 103 through the cellular network base station 102. The vehicle arrival information reporting data set is shown in table two, and mainly records a route number, a vehicle number, a station number, a time point, and the like, and the cloud computing server 103 converts the vehicle arrival information into station arrival travel time information (shown in table three), and stores the information in the cloud history database 105. For example, the time when the vehicle number 1 departs from station 1 is 2014/4/114: 46:28, and arrives at station 2 at 2014/4/114: 53:31, so the travel time from station 1 to station 2 is 423 seconds; and the time when the vehicle number 2 departs from station 1 is 2014/4/119: 32:22 and arrives at station 2 at 2014/4/119: 40:13, so the travel time from station 1 to station 2 is 471 seconds.
When the vehicle with the number 10001 travels to the station 6 (that is, the cloud server 103 knows the station-to-station travel time between the station 1 and the station 6), a station arrival time prediction system client device queries the cloud computing server 103 about the station arrival time of the station 12 with the route number 1 (that is, predicts the station-to-station travel time from the station 6 to the station 12 and converts the station arrival time into the station 12 arrival time). At this time, the cloud computing server 103 may use the data in the cloud historical database 105 (i.e., the station-to-station travel time information of the journey numbers 1 and 2, as shown in table four) as the data of the stochastic neural network group algorithm in the training phase to establish the stochastic neural network group, and use the algorithm to predict the station-to-station time.
Watch-to-station location information
Vehicle arrival information
Table three station to station travel time information
Training phase data of table four random neural network group algorithm
The method for predicting the arrival time based on the stochastic neural network group has the method flows as shown in fig. 2 and fig. 3. The method mainly comprises 2 stages: (a) a training phase and (b) a performance and testing phase.
The training phase mainly includes 4 steps, which are step S201: setting a random neural network group algorithm parameter value; s202: reading the travel time between each station in the historical database; s203: randomly generating m neural network models; and S208: and after filtering out the neural network models with the accuracy lower than the threshold value, remaining k neural network models.
The implementation and test phases mainly include 3 steps, which are S301: acquiring real-time travel time between each station and each station or test data in a test stage; s302: inputting data into the filtered k neural network models, and predicting station-to-station travel time; and S306: and obtaining the predicted station-to-station travel time, and converting the station-to-station travel time into the arrival time of the target station.
In the training phase, relevant parameter values of the stochastic neural network group algorithm are first set by the developer of the arrival time prediction system (step S201). For example, a total of 10 neural network models (i.e., m is 10), a maximum number of hidden layers in the neural network models is 5 (i.e., hmax is 5), a maximum number of neurons in each hidden layer in the neural network models is 7 (i.e., cmax is 7), a proportion of training data of the training neural network models to the total training-stage data is 60% (i.e., r% is 60%), and a correctness threshold value is 0.945 (i.e., wthreshold is 0.945 ═ 94.5%), and then 10 neural network models are generated according to the parameter values to predict arrival time.
In step S202, the time when the vehicle arrives at each station is obtained from the cloud history database, and is converted into the travel time between stations, as shown in table four. Since in the present embodiment, the vehicle to be predicted travels to the station 6, the arrival time at the station 12 is to be predicted, and the arrival time data between the stations 1 to 6 are known to be collected 1 2t,t, 3 4 5 6t,t,t,tConverts into a travel time data set from station to station 2 1t-t|,| 3 2t-t|,| 4 3t-t|,| 5 4t-t|,| 6 5t-t| and used for predicting the travel time from station 6 to station 12 (i.e. the travel time of target output is |) 12 6t-t|). In thatCollecting travel time data { [ 1 ] 2 1t-t|,| 3 2t-t|,| 4 3t-t|,| 5 4t-t|,| 6 5t-tRespectively named parameter names { x1, x2, x3, x4, x5}, and the travel time of the target output 12 6t-tI is named parameter name y.
In the step S203 of randomly generating m neural network models, the method further includes the step S204: training data and validation data are generated. Specifically, the invention randomly generates 10 neural network models according to the random neural network group algorithm parameter values set by the arrival time prediction system developer, and sets the maximum number of hidden layers in the neural network models to be 5 and the maximum number of neurons in each hidden layer in the neural network models to be 7, that is, the number of hidden layers in each neural network model will be between 0 and 5, and the number of neurons in each hidden layer will be between 0 and 7, and the embodiment of generating the result is shown in table five (step S205). The hidden layer of the neural network model 1 is 1 layer, and the number of neurons in the hidden layer is 2 (as shown in fig. 4); the hidden layer of the neural network-like model 2 is 2 layers, the number of neurons of the hidden layer 1 is 3, and the number of neurons of the hidden layer 2 is 4; by analogy, 10 neural network models can be obtained. In addition, since the TRaining Data number of the TRaining neural network model accounts for 60% of the total number of the TRaining Stage Data, taking table four as an example, the total number of the TRaining Stage Data is 10000, 6000 random strokes of the TRaining neural network model are taken out for learning and using as the TRaining neural network model for each neural network model, and the remaining 4000 strokes of TDTRS (Testing Data in TRaining Stage, test Data in TRaining Stage) are respectively used for verifying each neural network model in the TRaining Stage. In this step, 6000 sets of data obtained by each neural network-like model are generated randomly, and each neural network-like model obtains a different set of data to train and learn.
TABLE V stochastic neural network groups
Step S206: training and learning the neural network model. In the present embodiment, 6000 data are respectively input into the 10 neural network-like models for training and learning, and the following description uses the neural network-like model 1 (as shown in fig. 4) as an example, where the 6000 data in the neural network-like model 1 are a data combination including the route number 1 and not including the route number 10000, and the following description describes the training and learning of the neural network-like model 1.
Step i: the weights for each neuron are randomly generated, along with the constant terms for the hidden and output layer neurons, as shown in table six.
Weights of respective neurons of table six types of neural network model 1, and constant terms of neurons of hidden layer and output layer
| w1,6 | w2,6 | w3,6 | w4,6 | w5,6 | w1,7 | w2,7 | w3,7 | w4,7 | w5,7 | w6,8 | w7,8 | 6 | 7 | 8 |
| 0.7 | 0.7 | 0.2 | 0.1 | 0.6 | 0.1 | 0.8 | 0.5 | 0.3 | 1.0 | 0.6 | 0.6 | 0.8 | 0.7 | 0.3 |
Step ii: 6000 strokes of data are input into the neural network model 1 one by one, and the route number 1 is taken as an example below. Firstly normalizing the data to a value between 0 and 1,thus, the data in all examples are less than 5000, so the results are normalized by dividing by 5000 as shown in table seven. Then, the output signal of each hidden layer neuron is calculated according to the input signal, wherein the present embodiment adopts Logistic distribution (i.e. Logistic distribution)
) The output signal is calculated in the following manner.
Table seven normalized Path number 1 values
A neuron 6:
total input signal:
converting the output signal:
and 7, the neuron: total input signal:
converting the output signal:
step iii: and calculating the output signal of the neuron of the output layer according to the output signal of the hidden layer.
The neuron 8:
total input signal:
converting the output signal:
step iv: the error term of the output value (i.e., 0.759554) is compared to the true value (i.e., 0.7796).
step v: and feeding the error items back to the hidden layer, and respectively calculating the error items of neurons of the hidden layer.
step vi: the respective neuron weights and the constant term are updated in accordance with the neuron error term, and the learning rate σ is set to 0.8 in the present embodiment.
Step vii: and (5) repeating the steps ii to vi, inputting each data into the neural network model for learning, until the difference between the output signal of the round and the output signal of the previous round is lower than a threshold value othhold (in the example, othhold is set to be 0.01), converging and completing learning, and determining each neuron weight and constant term of the neural network model.
The above-mentioned training and learning process of the neural network-like model 1, and accordingly, other neural network-like models (i.e. the neural network-like model 2 to the neural network-like model 10) are trained at the same time, which can support parallel operation. After the training is finished, the steps ii to iii can be repeated when the travel time between the station 6 and the station 12 is predicted, the test data or the real-time data is used as an input signal, and the output signal is the predicted travel time value. The predicted travel time value generated by the neural network-like model needs to be normalized and reduced, so that the travel time second number can be obtained, for example: the output signal is 0.759554, which is multiplied by 5000, and the travel time is 3797.769233 seconds.
Step S207: and verifying and weighting the neural network model. After training and learning of all the neural network models are completed, the remaining 4000 pieces of data can be used for verifying each neural network model, and the average accuracy is calculated to serve as the weight of each neural network model. Taking the neural network-like model 1 as an example, all the test data in the training stage is input into the trained neural network-like model 1, and the steps ii to iii are repeated, so that the accuracy can be calculated. For example, when the route number 10000 is an input signal, the normalized numerical value is shown in table eight, and the predicted value is 0.75986369, and then the predicted value is multiplied by 5000 to 3799.318449, so that the accuracy is 1- (| true value-predicted value |/true value) ═ 1- (|3939-3799.318449|/3939) ═ 96.45%; by analogy, the average accuracy of the Test Data (TDTRS) in the 4000 training phases can be calculated, which in this example is 93.23%. In this embodiment, the average accuracy rates corresponding to the 10 neural network-like models are 93.23%, 94.90%, 94.03%, 93.57%, 94.61%, 93.52%, 94.93%, 95.21%, 94.48%, and 94.45%, respectively, as shown in table nine.
Table eight normalized distance number 10000 values
TABLE nine average accuracy for each neural network model
Step S208: and after filtering out the neural network models with the accuracy lower than the threshold value, remaining k neural network models. This step will analyze the average accuracy of each neural network model and filter out the values below the accuracy threshold wthreshold (i.e. 94.5% set in this embodiment), wherein 6 neural network models 1, 3, 4, 6, 9, 10, etc. will be filtered out, leaving 4 neural network models and their weights for the implementation and testing stages.
TABLE Ten, filtered neural network model and weighted values thereof
In step S301, in the execution and test phase, real-time vehicle arrival information is input to the trained stochastic neural network group, and arrival time prediction is performed. For example, when the arrival time prediction system client device is going to inquire the arrival time of the arrival station 12 at 2014/5/311: 59:00, the arrival time of the station 1 to the station 6 and the travel time between the stations (as shown in table eleven) are taken as the input data of the stochastic neural network group (as shown in table twelve), and the target predicted value of the travel time from the station 6 to the station 12 is obtained.
TABLE eleven vehicle arrival information
Twelve-station to-station travel time information
In addition, the arrival time prediction system developer at this stage can also collect historical data as Test Data (TDTES) in the test stage, and obtain the travel time between each station of each journey number as the input value of the stochastic neural network group so as to analyze and optimize the stochastic neural network group.
In step S302, data is input to the filtered k neural network-like models, and the station-to-station travel time is predicted. As shown in fig. 5, after the input data is obtained, the data can be used as input signals for each of the filtered neural network-like models (i.e., the neural network- like models 2, 5, 7, and 8, as shown in table ten), and the travel times are predicted to be 3766.607 seconds, 3857.98 seconds, 3661.828 seconds, and 3724.095 seconds by the neural network- like models 2, 5, 7, and 8, respectively (step S303), as shown in table thirteen. Finally, weighted averaging is performed according to the weight of each neural network model (steps S304 to S305) to obtain the travel time predicted value 3752.516552 seconds (i.e., [ 94.90% × 3766.607+ 94.61% × 3857.98+ 94.93% × 3661.828+ 95.21% × 3724.095]/[ 94.90% + 94.61% + 94.93% + 95.21% ] ═ 3752.516552).
Quasi-neural network model after filtering table thirteen and weighted value thereof
In step S306, the predicted station-to-station travel time is acquired and converted into the arrival time of the destination station. After the station-to-station travel time predicted value is obtained, the station-to-station travel time predicted value can be combined with the current station-to-station travel time predicted value to convert the station-to-station travel time predicted value into the arrival time of the target station. In this embodiment, the time point when the route number 10001 arrives at the station 6 is 2014/5/311: 58:46, and the predicted value of the travel time from the station 6 to the station 12 is 3752.516552 seconds, so that the predicted arrival time of the station 12 is 2014/5/313: 01:19, and the information is returned to the client device of the arrival time prediction system.
As an example of practical application to a passenger carrier, data of a passenger carrier a is proved, and data of the whole month in 3 months in 2014 are collected altogether, wherein 2956 runs are included, 40 road segments are covered in an experimental environment, and different data exploration algorithms are respectively adopted to test the accuracy of the data, including rogue Regression (LR), conventional Back-Propagation Neural Network (BPNN), and Random Neural Network (RNN) provided by the present invention.
TABLE fourteen comparison of the Performance of the present invention with other data exploration methods
| Method of producing a composite material | Accuracy rate |
| Average value method for historical data | 73.79％ |
| Rogis regression | 77.43％ |
| Inverse transmission type neural network | 77.88％ |
| The invention | 78.22％ |
In summary, the arrival time prediction system and method based on the stochastic neural network group of the present invention collects the travel time between stations of each road segment and each time period, and provides a novel stochastic neural network group to analyze the travel time data set, establish a plurality of neural network models to avoid the influence of extreme values, and comprehensively consider the prediction results of the neural network models to improve the prediction accuracy, so as to predict the arrival time of the bus to be taken by the user, and provide the predicted arrival time to the user as a reference.
The foregoing is by way of example only, and not limiting. It is intended that all equivalent modifications or variations without departing from the spirit and scope of the present invention shall be included in the appended claims.
[ notation ] to show
100: station board
101: vehicle-mounted terminal equipment
102: cellular network base station
103: cloud computing server
104: cloud computing machine room
105: cloud historical database
106: arrival time prediction system client device
S201-207, S301-S306: step (ii) of
1-8: neural network-like model
Based on the same inventive concept, the embodiment of the present invention further provides a mobile terminal, and since the method corresponding to the mobile terminal in fig. 3 is a method for starting the mobile terminal in the embodiment of the present invention, the implementation of the method in the embodiment of the present invention may refer to the implementation of the system, and repeated details are not repeated.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (3)
1. A method for predicting arrival time, the method comprising:
setting a plurality of random neural network group algorithm parameter values;
reading a plurality of travel times between stations in a historical database, and taking the set of the travel times as training data;
randomly generating m neural network-like models, wherein each of the neural network-like models randomly acquires r% of the training data for training and learning use and residual data of the training data for verification use, wherein the residual data accounts for (100% -r%) of the training data, and the r% of the training data acquired by each of the neural network-like models is different from each other, wherein each of the neural network-like models receives r% of the training data, generates a predicted travel time based on the residual data after convergence is reached, and the predicted travel time is compared with a correct travel time to acquire a correct rate of each of the neural network-like models;
filtering out the neural network models with the accuracy rate lower than a threshold value, and then remaining k neural network models, wherein the threshold value of the accuracy rate is 0.945;
obtaining a plurality of test data in the plurality of travel times or test stages between stations in real time, wherein when a user wants to inquire the arrival time of a vehicle at a station n and the vehicle is located at the station n-i, the plurality of travel times or test data comprise | t2-t1|,|t3-t2|,...,|tn-i-tn-i-1L where tkRepresenting the arrival time of the vehicle at the station k;
inputting the plurality of travel times or test data into k filtered neural network-like models, and predicting a plurality of station-to-station travel times from each k neural network-like models;
and after the plurality of predicted station-to-station travel times are obtained by each k neural network models, carrying out weighted average on the plurality of predicted station-to-station travel times according to the accuracy of each k neural network models so as to convert the plurality of predicted station-to-station travel times into the arrival time of the target station.
2. The method of claim 1, wherein the plurality of random neural network group algorithm parameter values include a number of neural network-like models, a maximum number of hidden layers in a neural network-like model, a maximum number of neurons per hidden layer in a neural network-like model, a ratio of a number of training data for training a neural network-like model to a total number of training phase data, and a correctness threshold value.
3. The method of claim 2, wherein the step of filtering out k neural network models with a correctness rate lower than a threshold value comprises:
and comparing the accuracy of the m randomly generated neural network-like models with the accuracy threshold value, and excluding the neural network-like models lower than the accuracy threshold value.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW103134687A TWI522974B (en) | 2014-10-06 | 2014-10-06 | Arrival time prediction system and method |
| TW103134687 | 2014-10-06 | ||
| CN201510094269.5A CN104715630B (en) | 2014-10-06 | 2015-03-03 | Arrival time prediction system and method |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510094269.5A Division CN104715630B (en) | 2014-10-06 | 2015-03-03 | Arrival time prediction system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106022541A CN106022541A (en) | 2016-10-12 |
| CN106022541B true CN106022541B (en) | 2020-12-01 |
Family
ID=53414915
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510094269.5A Expired - Fee Related CN104715630B (en) | 2014-10-06 | 2015-03-03 | Arrival time prediction system and method |
| CN201610472352.6A Expired - Fee Related CN106022541B (en) | 2014-10-06 | 2015-03-03 | Arrival time prediction method |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510094269.5A Expired - Fee Related CN104715630B (en) | 2014-10-06 | 2015-03-03 | Arrival time prediction system and method |
Country Status (2)
| Country | Link |
|---|---|
| CN (2) | CN104715630B (en) |
| TW (1) | TWI522974B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI591991B (en) * | 2016-01-29 | 2017-07-11 | Chunghwa Telecom Co Ltd | System and method for pre-signing vouchers for forecasting requests for traffic |
| CN105679021B (en) * | 2016-02-02 | 2018-11-06 | 招商局重庆交通科研设计院有限公司 | Journey time fusion forecasting and querying method based on traffic big data |
| CN105513362B (en) * | 2016-02-03 | 2017-11-03 | 东南大学 | A kind of bus platform adjacent area bus running state evaluation verification method |
| CN105489051A (en) * | 2016-02-15 | 2016-04-13 | 青岛海信电器股份有限公司 | Bus station arrival prompting method, mobile terminal and cloud server |
| CA3030580C (en) * | 2017-04-27 | 2021-06-29 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for determining an estimated time of arrival |
| WO2018227368A1 (en) * | 2017-06-13 | 2018-12-20 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending an estimated time of arrival |
| EP3635706B1 (en) | 2017-06-22 | 2021-11-17 | Beijing Didi Infinity Technology And Development Co., Ltd. | Methods and systems for estimating time of arrival |
| US11536582B2 (en) | 2017-08-10 | 2022-12-27 | Beijing Didi Infinity Technology And Development Co., Ltd. | System and method for estimating travel time and distance |
| EP3507783B1 (en) * | 2017-11-23 | 2021-11-10 | Beijing Didi Infinity Technology and Development Co., Ltd. | System and method for estimating arrival time |
| CN110361019B (en) * | 2018-04-11 | 2022-01-11 | 北京搜狗科技发展有限公司 | Method, device, electronic equipment and readable medium for predicting navigation time |
| CN109002722B (en) * | 2018-07-19 | 2021-06-29 | 郑州云海信息技术有限公司 | Method and system for estimating vulnerability scanning remaining time |
| CN109410580A (en) * | 2018-11-15 | 2019-03-01 | 山东管理学院 | A kind of real-time arrival time prediction technique of public transport and system |
| CN109920248B (en) * | 2019-03-05 | 2021-09-17 | 南通大学 | Bus arrival time prediction method based on GRU neural network |
| CN110459056A (en) * | 2019-08-26 | 2019-11-15 | 南通大学 | A kind of public transport arrival time prediction technique based on LSTM neural network |
| CN111667689B (en) * | 2020-05-06 | 2022-06-03 | 浙江师范大学 | Method, device and computer device for predicting vehicle travel time |
| US20220051115A1 (en) * | 2020-08-12 | 2022-02-17 | Allstate Insurance Company | Control Platform Using Machine Learning for Remote Mobile Device Wake Up |
| CN113706874B (en) * | 2021-10-28 | 2022-01-28 | 华清科盛(北京)信息技术有限公司 | Method and device for predicting arrival time of logistics vehicles in factory in real time and electronic equipment |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090062537A (en) * | 2007-12-13 | 2009-06-17 | (주)아이이지컨설팅 | A vehicles allocation control system |
| CN101630376B (en) * | 2009-08-12 | 2012-06-20 | 江苏大学 | Soft-sensing modeling method and soft meter of multi-model neural network in biological fermentation process |
| CN102081859B (en) * | 2009-11-26 | 2014-10-01 | 上海遥薇(集团)有限公司 | Control method of bus arrival time prediction model |
| CN102110365B (en) * | 2009-12-28 | 2013-11-06 | 日电（中国）有限公司 | Road condition prediction method and road condition prediction system based on space-time relationship |
| CN101794520B (en) * | 2010-03-15 | 2012-07-04 | 任兆玉 | Intelligent transportation system based on telecommunication technology and communication method thereof |
| CN102708701B (en) * | 2012-05-18 | 2015-01-28 | 中国科学院信息工程研究所 | System and method for predicting arrival time of buses in real time |
| CN103578267B (en) * | 2012-07-18 | 2015-12-16 | 北京掌行通信息技术有限公司 | Public transport based on public transport gps data is arrived at a station Forecasting Methodology and system |
| CN102854296B (en) * | 2012-08-30 | 2015-03-11 | 北京工业大学 | Sewage-disposal soft measurement method on basis of integrated neural network |
| CN102968670B (en) * | 2012-10-23 | 2016-08-17 | 北京京东世纪贸易有限公司 | The method and apparatus of prediction data |
| CN103885996A (en) * | 2012-12-24 | 2014-06-25 | 北京掌城科技有限公司 | Method of querying real-time bus arrival information |
| CN103236184B (en) * | 2013-04-03 | 2015-02-04 | 广州亿连达电子科技有限公司 | Method and device for automatic bus station report |
| TWM477647U (en) * | 2013-06-20 | 2014-05-01 | Trans Tech Integrate Company Ltd | RFID technique applied to bus stop for display of bus time estimation |
| CN103605725A (en) * | 2013-11-15 | 2014-02-26 | 中国联合网络通信集团有限公司 | Bus arrival time inquiring method, NFC terminal and server |
-
2014
- 2014-10-06 TW TW103134687A patent/TWI522974B/en not_active IP Right Cessation
-
2015
- 2015-03-03 CN CN201510094269.5A patent/CN104715630B/en not_active Expired - Fee Related
- 2015-03-03 CN CN201610472352.6A patent/CN106022541B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN104715630B (en) | 2017-05-10 |
| CN106022541A (en) | 2016-10-12 |
| TWI522974B (en) | 2016-02-21 |
| TW201614607A (en) | 2016-04-16 |
| CN104715630A (en) | 2015-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106022541B (en) | Arrival time prediction method | |
| Chen et al. | A short-term traffic prediction model in the vehicular cyber–physical systems | |
| CN104464291B (en) | Traffic flow predicting method and system | |
| CN106205114B (en) | A kind of Freeway Conditions information real time acquiring method based on data fusion | |
| CN102708701B (en) | System and method for predicting arrival time of buses in real time | |
| CN110348614B (en) | Method for obtaining passenger OD and method for predicting bus passenger flow | |
| CN110363984B (en) | Traffic flow prediction method and apparatus | |
| CN112288197B (en) | Intelligent scheduling method and device for station vehicles | |
| CN103678917A (en) | Bus real-time arrival time predicting method based on simulated annealing algorithm | |
| Wei et al. | Look-ahead insertion policy for a shared-taxi system based on reinforcement learning | |
| CN111582559A (en) | Method and device for estimating arrival time | |
| CN110570656B (en) | Method and device for customizing public transport line | |
| CN110866770A (en) | Prediction method and prediction system for vehicle maintenance scheme | |
| CN105243868A (en) | Bus arrival time forecasting method and device | |
| Habtie et al. | Artificial neural network based real-time urban road traffic state estimation framework | |
| CN105809290A (en) | Method and device for realizing logistic scheduling | |
| CN113763695A (en) | Dispatching method and system for automatic driving vehicle | |
| CN114723480B (en) | Passenger flow prediction method and cargo scheduling system for rural travel | |
| Duan et al. | A unified STARIMA based model for short-term traffic flow prediction | |
| CN111582527A (en) | Travel time estimation method and device, electronic equipment and storage medium | |
| Gao et al. | Learning-based open driver guidance and rebalancing for reducing riders’ wait time in ride-hailing platforms | |
| CN117216615A (en) | Method, system, medium and electronic equipment for identifying channeling behavior | |
| CN111489171B (en) | Riding travel matching method and device based on two-dimensional code, electronic equipment and medium | |
| Al Najada et al. | Highway cluster density and average speed prediction in vehicular ad hoc networks (vanets) | |
| CN112446531A (en) | Information processing method and information processing apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201201 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |