CN109489679A - A kind of arrival time calculation method in guidance path - Google Patents
A kind of arrival time calculation method in guidance path Download PDFInfo
- Publication number
- CN109489679A CN109489679A CN201811549331.5A CN201811549331A CN109489679A CN 109489679 A CN109489679 A CN 109489679A CN 201811549331 A CN201811549331 A CN 201811549331A CN 109489679 A CN109489679 A CN 109489679A
- Authority
- CN
- China
- Prior art keywords
- time
- section
- data
- neural network
- artificial neural
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
- G01C21/3492—Special cost functions, i.e. other than distance or default speed limit of road segments employing speed data or traffic data, e.g. real-time or historical
Abstract
The invention discloses the arrival time calculation methods in a kind of guidance path.Utilize the mass data in historical record, by way of Machine self-learning, to road speed limit, vehicle fleet size, lane quantity and the speed per hour data of user to user's speed per hour, and red time, green time, queuing vehicle fleet size and corresponding calculated and trained by time data, the interaction between above-mentioned factor is comprehensively considered, and analyze by the collective effect of time, arrival time is calculated and is optimized, computational accuracy is improved.
Description
Technical field
The invention belongs to communication navigation technical fields, more particularly, to the arrival time calculating side in a kind of guidance path
Method.
Background technique
With the development of modern society, voluntarily drive to go on a journey as more and more common phenomenon, and city vehicle quantity
It sharply increases, results in urban road congestion and frequently occur, directly contribute the waste of a large amount of travel time, go out line efficiency and significantly drop
The problems such as fuel consumption that is low, thus deriving, air pollution, road anger disease.Seriously reduce quality of the life, cause it is economical and
Social concern.
With network mobile terminals such as the rapid development of wireless communication and mobile calculation technique and mobile phone, tablet computers
Rapid development, trip airmanship obtained quick development, such as Gao De, Tencent, Baidu map have trip route choosing
The functions such as navigation are selected, have become the indispensable tool of people's trip at present.During navigation, path planning
Afterwards, particularly significant to estimating for the navigation arrival time of each path, it is the important references that user selects reliable guidance path
Foundation.At present planning or navigate remaining time determination, be according to each road calculated by the time and acquisition of summing
's.Mainly pass through road distance, user's speed per hour and road speed limit, jam situation, traffic lights number etc. above by the time
Condition of road surface comprehensively considers, and since the factor of consideration is quite a lot of, the side of estimation is all made of by the influence of time to each factor
Formula, there are certain error, the sum of Multiple factors will result in sizable error for the estimation of single factor, affect calculating essence
Degree.
Wherein artificial neural network (Artificial Neural Network, i.e. ANN) is that artificial intelligence field rises
Research hotspot.It is abstracted human brain neuroid from information processing angle, certain naive model is established, by a large amount of
Different networks is mutually formed by different connection types between node (or neuron).Each node on behalf is a kind of specific
Output function, referred to as excitation function (activation function).Connection between every two node all represents one for logical
Cross the weighted value of the connection signal, referred to as weight.The output of network is then according to the connection type of network, weighted value and excitation function
Difference and it is different.Artificial neural network is in pattern-recognition, intelligent robot, automatic control, predictive estimation, biology, medicine, warp
The fields such as Ji have successfully solved many practical problems, show good intelligent characteristic.But it is led at present in communication navigation
Domain there is no the report of application.
Summary of the invention
In order to solve the above technical problems, the present invention provides the arrival time calculation methods in a kind of guidance path.
The complete technical solution of the present invention includes:
A kind of arrival time calculation method in guidance path, which comprises the steps of:
(1) it is directed to current location and destination, section division is carried out to selected path, draws entire path according to node
It is divided into different 1~n of section, node includes: one of traffic lights, number of track-lines amount change location, section speed limit change position, section
Section is waited including normally travel section and traffic lights.
(2) being calculated by the time to each section:
Such as normally travel section 1, it is calculated in the following way and passes through time t1,
S is 1 length of section, V in formula1Speed per hour is calculated for the user.
The wherein calculating speed per hour V of user1It is calculated using the artificial nerve network model after training.Specifically: selection
Record is travelled in historical record, can be directed to certain a road section or a plurality of section, be extracted wherein the road speed limit in each section, vehicle number
Amount, lane quantity and the speed per hour data of user by extracting, filling a vacancy, being formed after smoothing and noise-reducing process database, and are based on this
Database establishes artificial neural network system, inputs road speed limit, vehicle fleet size, lane incremental data, and with this condition
User's speed per hour data, be trained using data of the artificial neural network system to input, with the artificial neural network to foundation
Network system optimizes, and using lane, vehicle fleet size and the speed limit data of artificial neural network input at this time after optimization, carries out
User's speed per hour V under this road conditions1It calculates.
Such as traffic lights section 2 traffic lights by the time, recorded by the traveling in selection historical record, can be with
For a certain traffic lights or multiple traffic lights, extract the wherein red time of each traffic lights, green time, queuing vehicle fleet size
And it is corresponding by time data, by extracting, filling a vacancy, being formed after smoothing and noise-reducing process database, and it is based on this database,
Artificial neural network system is established, the red time of each traffic lights, green time, the vehicle fleet size of queuing and corresponding are inputted
By time data, it is trained using data of the artificial neural network system to input, with the artificial neural network to foundation
System optimizes, using after optimization artificial neural network input red time at this time, green time, queuing vehicle number
Data are measured, are carried out through time t2It calculates.
(3) arrival time in guidance path can be obtained after being added in each section in step (3) by the time.
Pass through the fitting precision of Mean Square Error MSE metrics evaluation artificial neural network.Formula is as follows:
Wherein N0It is the group number of output data, Q is the group number of training data, and d is experimental data, and y is neural network output
Data.Trained target is correlation > 9, MSE less than 0.001.
The present invention compared with the existing technology the advantages of be: using the mass data in historical record, learnt by oneself by machine
The mode of habit, when to road speed limit, vehicle fleet size, lane quantity and the speed per hour data of user to user's speed per hour and red light
Between, green time, queuing vehicle fleet size and it is corresponding calculated and trained by time data, comprehensively considered above-mentioned
Interaction between factor, and analyze by the collective effect of time, arrival time is calculated and is optimized,
Improve computational accuracy.
Detailed description of the invention
Fig. 1 is the flow chart of presently disclosed method.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and detailed description.
A kind of arrival time calculation method in guidance path, includes the following steps:
(1) it is directed to current location and destination, section division is carried out to selected path, draws entire path according to node
It is divided into different 1~n of section, node includes: one of traffic lights, number of track-lines amount change location, section speed limit change position.
1 section of table divides example
As shown above, entire path is divided into five sections, wherein 1 lane quantity of section be 4, speed limit 80Km/h, and
There are railing or isolation strip to be isolated with the non-motorized lane in roadside, is indicated with 1.Section 2 is traffic light.3 lane quantity of section
It is 3, speed limit 60Km/h, and there is railing or isolation strip to be isolated with the non-motorized lane in roadside.Section 4 is traffic light.Road
5 lane quantity of section are 2, speed limit 40Km/h, and are isolated with the non-motorized lane in roadside without railing or isolation strip, are indicated with 0.
(2) being calculated by the time to each section can be obtained after addition entirely through the time.
Wherein for section 1, it is calculated in the following way and passes through time t1,
S is 1 length of section, V in formula1Speed per hour is calculated for the user.
The wherein calculating speed per hour V of user1It is calculated using the artificial nerve network model after training.Due to each user's
Driving habit is different, and some is got used to travelling at high speeds, and speed per hour is essentially close to road speed limit, and some is then safer,
In the speed per hour downward driving for being significantly lower than road speed limit.It is thus determined that the basic driving habit of each user, for passing through the time really
Surely it is even more important.Clearly for the influence by the time, vehicle is less for the vehicle fleet size travelled on section at present simultaneously
In the case of influence to user's speed per hour it is little, but when vehicle fleet size is more than certain threshold value, the driving that can significantly affect user is practised
It is used, than if any user more conservative drive manner can be taken when vehicle fleet size increases, while different user is to vehicle fleet size
Response it is also not identical.But under the premise of identical vehicle fleet size, the lane quantity being arranged on section at present is for passing through the time
Influence equally clearly, then also will affect the driving habit of user.Therefore, driving habit, lane quantity, vehicle fleet size
Influence to user's speed is coefficient, while also reciprocal effect between these three factors, is to take individually to estimate at present
It calculates, and the mode for increasing the time calculated by the time.But for there is presently no the factors for being directed to this three's reciprocal effect
To the coefficient effective way for passing through the time.
Due to that in navigational tool use process, can collect a large amount of user's trip historical record, the present invention then passes through this
The big data of collection, is trained using artificial nerve network model, and predicts that user's passes through the time in turn.
The traveling record in user's history record is selected, a plurality of section is selected, extracts wherein the road speed limit in each section, vehicle
The speed per hour data of quantity, lane quantity and user, being formed by extracting, filling a vacancy, after smoothing and noise-reducing process includes 1000
The database of record, and it is based on this database, artificial neural network system is established, road speed limit, vehicle fleet size, number of track-lines are inputted
Data, and user's speed per hour data with this condition are measured, wherein using 800 datas as training sample, using artificial mind
It is trained through data of the network system to input, to be optimized to the artificial neural network system of foundation, using 200 numbers
According to verifying sample is carried out, using lane, vehicle fleet size and the speed limit data of artificial neural network input at this time after optimization, carry out
User's speed per hour V under certain road conditions1Prediction.
A certain friendship can be directed to by the traveling record in selection historical record by the time for the traffic lights in section 2
Logical lamp or multiple traffic lights extract the wherein red time of each traffic lights, green time, the vehicle fleet size of queuing and corresponding
By time data, by extracting, filling a vacancy, being formed after smoothing and noise-reducing process database, data format example such as table 2, and it is based on
This database, establishes artificial neural network system, input the red time of each traffic lights, green time, queuing vehicle fleet size
And it is corresponding by time data, it is trained using data of the artificial neural network system to input, with the people to foundation
Artificial neural networks system optimizes, and utilizes red time, the green time, row of artificial neural network input at this time after optimization
The vehicle fleet size data of team, carry out through time t2It calculates.
2 intersection data format of table
Crossing | Red time (s) | Green time (s) | Queuing vehicle quantity | Pass through the time (s) |
1 | 135 | 45 | 10 | 27 |
2 | 45 | 45 | 20 | 124 |
3 | 120 | 20 | 5 | 65 |
4 | 45 | 45 | 11 | 20 |
Further, pass through the fitting precision of Mean Square Error MSE metrics evaluation artificial neural network.Formula is as follows:
Wherein N0It is the group number of output data, Q is the group number of training data, and d is experimental data, and y is neural network output
Data.Trained target is correlation > 9, MSE less than 0.001.It finds after study, by hidden layer neuron quantity from 3-15
Successively change and train network, finds hidden layer neuron quantity when 8, the MSE of the present inventor's artificial neural networks is in
Relative thereto, related coefficient (Regression) are in opposite high point, and network meets training objective at this time, and has relatively most
Good fitting precision.
The calculating time in remaining section with it is similarly as described above.
To calculating by the time for each section, whole arrival time can be obtained after addition.
The above is only presently preferred embodiments of the present invention, is not intended to limit the invention in any way, it is all according to the present invention
Technical spirit any simple modification to the above embodiments, change and equivalent structural changes, still fall within skill of the present invention
In the protection scope of art scheme.
Claims (2)
1. the arrival time calculation method in a kind of guidance path, which comprises the steps of:
(1) it is directed to current location and destination, section division is carried out to selected path, is divided into entire path according to node
Different 1~n of section, node include: one of traffic lights, number of track-lines amount change location, section speed limit change position, and section includes
Normally travel section and traffic lights wait section.
(2) being calculated by the time to each section:
Such as normally travel section 1, it is calculated in the following way and passes through time t1,
S is 1 length of section, V in formula1Speed per hour is calculated for the user.
The wherein calculating speed per hour V of user1It is calculated using the artificial nerve network model after training.Specifically: selection history note
Record is travelled in record, can be directed to certain a road section or a plurality of section, be extracted wherein the road speed limit in each section, vehicle fleet size, vehicle
Road quantity and the speed per hour data of user by extracting, filling a vacancy, being formed after smoothing and noise-reducing process database, and are based on this data
Artificial neural network system is established in library, inputs road speed limit, vehicle fleet size, lane incremental data, and use with this condition
Family speed per hour data are trained using data of the artificial neural network system to input, with the artificial neural network system to foundation
System optimizes, and using lane, vehicle fleet size and the speed limit data of artificial neural network input at this time after optimization, carries out this road
User's speed per hour V under the conditions of road1It calculates.
Such as the traffic lights in traffic lights section 2 can be directed to by the time by the traveling record in selection historical record
A certain traffic lights or multiple traffic lights, extract wherein the red time of each traffic lights, green time, queuing vehicle fleet size and
It is corresponding to be established by time data by extracting, filling a vacancy, being formed after smoothing and noise-reducing process database, and based on this database
Artificial neural network system, input the red time of each traffic lights, green time, queuing vehicle fleet size and corresponding pass through
Time data are trained, with the artificial neural network system to foundation using data of the artificial neural network system to input
Optimize, using after optimization artificial neural network input red time at this time, green time, queuing vehicle fleet size number
According to carrying out through time t2It calculates.
(3) arrival time in guidance path can be obtained after being added in a section in step (3) by the time.
2. the arrival time calculation method in a kind of guidance path as described in claim 1, which is characterized in that
Pass through the fitting precision of Mean Square Error MSE metrics evaluation artificial neural network.Formula is as follows:
Wherein N0It is the group number of output data, Q is the group number of training data, and d is experimental data, and y is neural network output data.
Trained target is correlation > 9, MSE less than 0.001.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811549331.5A CN109489679B (en) | 2018-12-18 | 2018-12-18 | Arrival time calculation method in navigation path |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811549331.5A CN109489679B (en) | 2018-12-18 | 2018-12-18 | Arrival time calculation method in navigation path |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109489679A true CN109489679A (en) | 2019-03-19 |
CN109489679B CN109489679B (en) | 2019-12-17 |
Family
ID=65710677
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811549331.5A Expired - Fee Related CN109489679B (en) | 2018-12-18 | 2018-12-18 | Arrival time calculation method in navigation path |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109489679B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111397631A (en) * | 2020-04-10 | 2020-07-10 | 上海安吉星信息服务有限公司 | Navigation path planning method and device and navigation equipment |
CN111582559A (en) * | 2020-04-21 | 2020-08-25 | 腾讯科技(深圳)有限公司 | Method and device for estimating arrival time |
WO2021051329A1 (en) * | 2019-09-19 | 2021-03-25 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for determining estimated time of arrival in online to offline services |
CN115218912A (en) * | 2021-12-10 | 2022-10-21 | 广州汽车集团股份有限公司 | Navigation duration estimation method and device, vehicle and equipment |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101436347A (en) * | 2008-12-09 | 2009-05-20 | 北京交通大学 | Prediction method for rapid road travel time |
CN101479733A (en) * | 2006-06-30 | 2009-07-08 | 微软公司 | Inferring road speeds for context-sensitive routing |
CN102081859A (en) * | 2009-11-26 | 2011-06-01 | 上海遥薇实业有限公司 | Control method of bus arrival time prediction model |
CN102682591A (en) * | 2011-03-16 | 2012-09-19 | 高德软件有限公司 | Method and device for acquiring travel time |
CN104217605A (en) * | 2013-05-31 | 2014-12-17 | 张伟伟 | Bus arrival time estimation method and device |
CN105270383A (en) * | 2014-05-30 | 2016-01-27 | 福特全球技术公司 | Vehicle speed profile prediction using neural networks |
CN106679685A (en) * | 2016-12-29 | 2017-05-17 | 鄂尔多斯市普渡科技有限公司 | Driving path planning method for vehicle navigation |
CN107305742A (en) * | 2016-04-18 | 2017-10-31 | 滴滴(中国)科技有限公司 | Method and apparatus for determining E.T.A |
CN108288096A (en) * | 2017-01-10 | 2018-07-17 | 北京嘀嘀无限科技发展有限公司 | Method and device for estimating journey time, model training |
US10127496B1 (en) * | 2017-11-23 | 2018-11-13 | Beijing Didi Infinity Technology And Development | System and method for estimating arrival time |
-
2018
- 2018-12-18 CN CN201811549331.5A patent/CN109489679B/en not_active Expired - Fee Related
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101479733A (en) * | 2006-06-30 | 2009-07-08 | 微软公司 | Inferring road speeds for context-sensitive routing |
CN101436347A (en) * | 2008-12-09 | 2009-05-20 | 北京交通大学 | Prediction method for rapid road travel time |
CN102081859A (en) * | 2009-11-26 | 2011-06-01 | 上海遥薇实业有限公司 | Control method of bus arrival time prediction model |
CN102682591A (en) * | 2011-03-16 | 2012-09-19 | 高德软件有限公司 | Method and device for acquiring travel time |
CN104217605A (en) * | 2013-05-31 | 2014-12-17 | 张伟伟 | Bus arrival time estimation method and device |
CN105270383A (en) * | 2014-05-30 | 2016-01-27 | 福特全球技术公司 | Vehicle speed profile prediction using neural networks |
CN107305742A (en) * | 2016-04-18 | 2017-10-31 | 滴滴(中国)科技有限公司 | Method and apparatus for determining E.T.A |
CN106679685A (en) * | 2016-12-29 | 2017-05-17 | 鄂尔多斯市普渡科技有限公司 | Driving path planning method for vehicle navigation |
CN108288096A (en) * | 2017-01-10 | 2018-07-17 | 北京嘀嘀无限科技发展有限公司 | Method and device for estimating journey time, model training |
US10127496B1 (en) * | 2017-11-23 | 2018-11-13 | Beijing Didi Infinity Technology And Development | System and method for estimating arrival time |
Non-Patent Citations (1)
Title |
---|
蒋渭忠 等,: ""基于BP神经网络技术的车辆通行时间预测研究"", 《微型机与应用》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021051329A1 (en) * | 2019-09-19 | 2021-03-25 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for determining estimated time of arrival in online to offline services |
CN111397631A (en) * | 2020-04-10 | 2020-07-10 | 上海安吉星信息服务有限公司 | Navigation path planning method and device and navigation equipment |
CN111582559A (en) * | 2020-04-21 | 2020-08-25 | 腾讯科技(深圳)有限公司 | Method and device for estimating arrival time |
CN111582559B (en) * | 2020-04-21 | 2024-02-20 | 腾讯科技(深圳)有限公司 | Arrival time estimation method and device |
CN115218912A (en) * | 2021-12-10 | 2022-10-21 | 广州汽车集团股份有限公司 | Navigation duration estimation method and device, vehicle and equipment |
CN115218912B (en) * | 2021-12-10 | 2023-11-21 | 广州汽车集团股份有限公司 | Navigation duration prediction method, navigation duration prediction device, vehicle and navigation duration prediction equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109489679B (en) | 2019-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108761509B (en) | Automobile driving track and mileage prediction method based on historical data | |
CN105913661B (en) | A kind of express highway section traffic state judging method based on charge data | |
CN101965601B (en) | Driving support device and driving support method | |
CN109489679A (en) | A kind of arrival time calculation method in guidance path | |
CN107563566B (en) | Inter-bus-station operation time interval prediction method based on support vector machine | |
CN110298500A (en) | A kind of urban transportation track data set creation method based on taxi car data and city road network | |
CN108364467A (en) | A kind of traffic information prediction technique based on modified decision Tree algorithms | |
CN107490384B (en) | Optimal static path selection method based on urban road network | |
CN111210612B (en) | Method for extracting bus route track based on bus GPS data and station information | |
CN110836675B (en) | Decision tree-based automatic driving search decision method | |
CN110274609B (en) | Real-time path planning method based on travel time prediction | |
CN109686090B (en) | Virtual traffic flow calculation method based on multi-source data fusion | |
CN106710208A (en) | Traffic state acquisition method and device | |
CN110176139A (en) | A kind of congestion in road identification method for visualizing based on DBSCAN+ | |
CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
CN106205156A (en) | A kind of crossing self-healing control method for the sudden change of part lane flow | |
CN105447504A (en) | Traffic mode behavior recognition method and corresponding recognition model construction method | |
CN112767683B (en) | Path induction method based on feedback mechanism | |
CN108961758A (en) | A kind of crossing broadening lane detection method promoting decision tree based on gradient | |
CN106097717A (en) | The signalized intersections average transit time method of estimation merged based on two class floating car datas | |
CN106227859A (en) | The method identifying the vehicles from gps data | |
CN111191816A (en) | System for identifying travel time chain of urban rail transit passengers | |
CN112036757A (en) | Parking transfer parking lot site selection method based on mobile phone signaling and floating car data | |
CN112784000A (en) | Passenger searching method based on taxi track data | |
CN112967493A (en) | Neural network-based prediction method for vehicle passing intersection travel time |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191217 Termination date: 20201218 |
|
CF01 | Termination of patent right due to non-payment of annual fee |