CN110487294A - Intelligent path planning system and method based on weather cloud atlas - Google Patents
Intelligent path planning system and method based on weather cloud atlas Download PDFInfo
- Publication number
- CN110487294A CN110487294A CN201910808259.1A CN201910808259A CN110487294A CN 110487294 A CN110487294 A CN 110487294A CN 201910808259 A CN201910808259 A CN 201910808259A CN 110487294 A CN110487294 A CN 110487294A
- Authority
- CN
- China
- Prior art keywords
- cloud
- thunder
- atlas
- weather
- path
- 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.)
- Pending
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
- G01C21/3407—Route searching; Route guidance specially adapted for specific applications
- G01C21/3415—Dynamic re-routing, e.g. recalculating the route when the user deviates from calculated route or after detecting real-time traffic data or accidents
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Environmental & Geological Engineering (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Atmospheric Sciences (AREA)
- Environmental Sciences (AREA)
- Automation & Control Theory (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Navigation (AREA)
Abstract
The present invention relates to a kind of intelligent path planning systems and method based on weather cloud atlas, and the system comprises thunder cloud identification modules；Thunder cloud prediction module；Stroke obtains module；Path planning module, for there are when driving plan of travel in the routing of user, plan departure time, departure place, destination and arrival time corresponding to the driving plan of travel are obtained, and driving path planning is carried out according to the driving plan of travel, obtains the first planning path；Path adjusts module, for judging in first planning path with the presence or absence of the location point for being less than or equal to pre-determined distance at a distance from the travel path of thunder cloud, if it is, first planning path is then adjusted, so that each location point is all larger than pre-determined distance at a distance from the travel path of the thunder cloud in first planning path.The present invention is based on weather cloud atlas to carry out path planning to the driving plan of travel in user's stroke in advance, user-friendly.
Description
Technical field
The present invention relates to Path Planning Technique field, in particular to a kind of intelligent path planning systems based on weather cloud atlas
And method.
Background technique
The trip of oneself is carried out with the development of society, the trip of people rely increasingly upoies high-precision digital map navigation
Path planning.When navigation system carries out path planning, can be joined according to road type, road toll situation, road condition etc.
Number synthesis considers that selection is appropriate to the trip route of user.However, in practical applications, user is during trip, it is also possible to
It is influenced by a lot of other factors, such as weather conditions etc., if it is possible to be used well using the prediction help of weather conditions
Family avoid in advance it is some it is possible that poor state of weather path, then be advantageous to the trip of user.
Summary of the invention
The present invention provides a kind of intelligent path planning systems and method based on weather cloud atlas, existing its object is to overcome
There is the defects of technology, based on the travel path of weather cloud atlas look-ahead thunder cloud, user is helped according to the stroke of user in advance
The trip route of thunder cloud is avoided in arrangement, is conducive to user's safety.
To achieve the goals above, the present invention has following constitute:
The intelligent path planning system based on weather cloud atlas, comprising:
Thunder cloud identification module is analyzed in the weather cloud atlas for obtaining weather cloud atlas with the presence or absence of thunder cloud；
Thunder cloud prediction module when for recognizing thunder cloud in the weather cloud atlas, calculates in the weather cloud atlas
The movement velocity and the direction of motion of thunder cloud, and it is pre- according to the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas
Survey the travel path of the thunder cloud in the first preset time period；
Stroke obtains module, and the routing in the first preset time period for obtaining user judges the stroke peace
Whether there is in row has driving plan of travel；
Path planning module, for, there are when driving plan of travel, obtaining described drive out in the routing of user
Row plan corresponding plan departure time, departure place, destination and arrival time, and according to the driving plan of travel into
Row driving path planning, obtains the first planning path；
Path adjusts module, for judging in first planning path with the presence or absence of the travel path with the thunder cloud
Distance be less than or equal to the location point of pre-determined distance, if it is, adjustment first planning path, so that first planning
Each location point is all larger than pre-determined distance at a distance from the travel path of the thunder cloud in path.
Optionally, the system also includes identification model training modules, for training thunder cloud identification model, the thunderstorm
Cloud identification model is to be trained using cloud atlas area image training set to the deep learning model based on convolutional neural networks
Arrive, in the cloud atlas area image training set include it is multiple include thunder cloud cloud atlas area image and it is multiple do not include thunder
The cloud atlas area image of sudden and violent cloud, and the label of thunder cloud is made whether in each cloud atlas area image, the thunder cloud identification
The input of model is the cloud atlas area image of pre-set dimension, is exported as there are the probability of thunder cloud in the cloud atlas region；
The thunder cloud identification module is identified using following steps whether there is thunder cloud in the weather cloud atlas:
The weather cloud atlas is divided into the cloud atlas region of multiple pre-set dimensions by the thunder cloud identification module, by multiple institutes
It states cloud atlas region and sequentially inputs trained thunder cloud identification model, if there are the probability of thunder cloud to be greater than in a cloud atlas region
Predetermined probabilities threshold value then judges that there are thunder clouds in the cloud atlas region；
The thunder cloud identification module judges that there are thunders with the presence or absence of at least one described cloud atlas region in a weather cloud atlas
Sudden and violent cloud, if it is, determining that there are thunder clouds in the weather cloud atlas.
Optionally, the determination thunder cloud identification module determines in the weather cloud atlas there are after thunder cloud, goes back
Include the following steps:
There are each cloud atlas regions of the thunder cloud whether there is adjacent cloud atlas for the thunder cloud identification module judgement
Region then records the thunder cloud in adjacent cloud atlas region if there is there are the thunder clouds in adjacent cloud atlas region
It for the same thunder cloud, and is numbered for each thunder cloud, the number of the thunder cloud in adjacent cloud atlas region is same.
Optionally, the thunder cloud prediction module calculates the movement speed of thunder cloud in the weather cloud atlas using following steps
Degree and the direction of motion:
The thunder cloud prediction module obtains multiple weather cloud atlas in the second preset time period in the past, obtains each number
Position of the corresponding thunder cloud in every weather cloud atlas；
According to the position of acquisition time corresponding to every weather cloud atlas and each thunder cloud in every weather cloud atlas, in advance
Survey the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas.
Optionally, path adjustment module be also used to judge in the departure place and destination of the driving plan of travel be
It is no at least one at a distance from the travel path of the thunder cloud be less than or equal to pre-determined distance location point；
If only departure place is less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, it is determined that described
Arrival time in the travel path of thunder cloud with the departure place apart from nearest point, calculate arrival time of nearest point with
The time difference of departure time determines the risk class that sets out according to the mapping relations of preset time difference and risk class, will
The risk class that sets out is sent to user；
If being only purposefully less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, it is determined that described
Arrival time in the travel path of thunder cloud with the destination apart from nearest point, calculate arrival time of nearest point with
The time difference of arrival time determines arrival risk class according to the mapping relations of preset time difference and risk class, will
The arrival risk class is sent to user；
If departure place and destination are less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, divide
Not Ji Suan arrival time and departure time in the travel path of thunder cloud with departure place apart from nearest point first time it is poor
The second time difference of arrival time and arrival time in the travel path of value and thunder cloud with destination apart from nearest point,
It is determined respectively according to first time difference and the second time difference and sets out risk class and reach risk class, and be sent to use
Family.
Optionally, described set out risk class and/or is reached risk class and is sent to user by path adjustment module
Later, it is also used to receive the driving plan of travel modification information of user, is modified according to the driving plan of travel modification information of user
The routing of user.
The embodiment of the present invention also provides a kind of intelligent paths planning method based on weather cloud atlas, using described based on day
The intelligent path planning system of gas cloud figure, described method includes following steps:
Weather cloud atlas is obtained, is identified in the weather cloud atlas with the presence or absence of thunder cloud；
If there is thunder cloud, then the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas are calculated；
The institute in the first preset time period is predicted according to the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas
State the travel path of thunder cloud；
The routing in the first preset time period of user is obtained, judging, which whether there is in the routing, driving
Plan of travel；
If there is driving plan of travel, then obtains the plan travel time corresponding to the driving plan of travel, sets out
Ground and destination；
Driving path planning is carried out according to the driving plan of travel, obtains the first planning path；
Judge pre- with the presence or absence of being less than or equal at a distance from the travel path of the thunder cloud in first planning path
If the location point of distance；
If it is, adjustment first planning path so that in first planning path each location point with it is described
The distance of the travel path of thunder cloud is all larger than pre-determined distance.
Optionally, the step of the method also includes training thunder cloud identification models, the trained thunder cloud identification model,
Include the following steps:
Acquire it is multiple include thunder cloud cloud atlas area image and it is multiple do not include thunder cloud cloud atlas area image, and
It is made whether the label of thunder cloud in each cloud atlas area image, cloud atlas administrative division map is added in the cloud atlas area image after label
As training set；
The deep learning model based on convolutional neural networks is trained using the cloud atlas area image training set
To shown thunder cloud identification model, the input of the thunder cloud identification model is the cloud atlas area image of pre-set dimension, exports and is
There are the probability of thunder cloud in the cloud atlas region；
It whether there is thunder cloud in the identification weather cloud atlas, include the following steps:
The weather cloud atlas is divided into the cloud atlas region of multiple pre-set dimensions by the thunder cloud identification module, by multiple institutes
It states cloud atlas region and sequentially inputs trained thunder cloud identification model, if there are the probability of thunder cloud to be greater than in a cloud atlas region
Predetermined probabilities threshold value then judges that there are thunder clouds in the cloud atlas region；
The thunder cloud identification module judges that there are thunders with the presence or absence of at least one described cloud atlas region in a weather cloud atlas
Sudden and violent cloud, if it is, determining that there are thunder clouds in the weather cloud atlas.
Using the intelligent path planning system and method based on weather cloud atlas in the invention, have following beneficial to effect
Fruit:
The present invention is based on the travel paths of weather cloud atlas look-ahead thunder cloud, help user to pacify in advance according to the stroke of user
Row avoids the trip route of thunder cloud, is conducive to user's safety；The present invention is based on the convolutional neural networks moulds of deep learning
Thunder cloud in type automatic identification weather cloud atlas, and according to the traveling of the dynamic prediction thunder cloud of thunder cloud in multiple weather cloud atlas
Direction and travel speed, so that thunder cloud travel path quick predict is realized, to improve processing speed；Further, this hair
It is bright that plan of travel is directly got from the routing of user, the path for avoiding insecurity factor is arranged for user in advance, is mentioned
High user's safety, and facilitate user to make strain early in the case where being likely encountered bad weather and arrange.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the intelligent path planning system based on weather cloud atlas of one embodiment of the invention；
Fig. 2 is the schematic diagram based on weather cloud atlas revised planning path of one embodiment of the invention；
Fig. 3 is the flow chart of the intelligent paths planning method based on weather cloud atlas of one embodiment of the invention；
Fig. 4 is the flow chart of the training thunder cloud identification model of one embodiment of the invention.
Appended drawing reference:
M100 thunder cloud identification module
M200 thunder cloud prediction module
M300 stroke obtains module
M400 path planning module
The path M500 adjusts module
M600 model training module
Specific embodiment
It is further to carry out combined with specific embodiments below in order to more clearly describe technology contents of the invention
Description.
The technical issues of in order to solve in the prior art, the present invention provides a kind of, and the intelligent path based on weather cloud atlas is advised
The system of drawing.As shown in Figure 1, in an embodiment of the present invention, the intelligent path planning system based on weather cloud atlas includes:
Thunder cloud identification module M100 is analyzed in the weather cloud atlas for obtaining weather cloud atlas with the presence or absence of thunder cloud；
Thunder cloud is to be developed by cumulonimbus monomer or be composed of multiple cumulonimbus in different stages of development.Thunder cloud must be
It could be formed under conditions of unstable atmospheric stratification, abundant steam and enough impact forces.It each thunderstorm cell and developed
Journey will substantially undergo development, maturation and dissipation three phases.About more than ten kilometers of its horizontal extent, thick several kilometers to more than ten of cloud public
In, duration dozens of minutes belongs to small scale weather system.The appearance of thunder cloud, it is possible that the rainfall of big degree, right
The safety of user causes a hidden trouble；
Thunder cloud prediction module M200 when for recognizing thunder cloud in the weather cloud atlas, calculates the day gas cloud
The movement velocity and the direction of motion of thunder cloud in figure, and according to the movement velocity of thunder cloud in the weather cloud atlas and movement side
To the travel path of prediction thunder cloud in the first preset time period；
Stroke obtains module M300, and the routing in the first preset time period for obtaining user judges the row
Whether there is in journey arrangement has driving plan of travel；First preset time period, which can according to need, to be selected, can also by with
Family is set oneself, such as routing in the routing in three days futures of setting, or one week future of setting etc., from
It is dynamic to obtain plan of travel from the routing of user, it is operated again without user, and it is longer out that the time may be implemented
The schedule ahead of row plan；In this embodiment it is possible to from the calendar of the mobile terminal of user, memorandum or routing APP
In get the routing of user, however, the present invention is not limited thereto kind mode；
Path planning module M400, for, there are when driving plan of travel, being opened described in acquisition in the routing of user
Plan departure time, departure place, destination and arrival time corresponding to vehicle plan of travel, and gone on a journey and counted according to the driving
It draws and carries out driving path planning, obtain the first planning path；The first planning path can be navigation system according to existing herein
Path planning algorithm comprehensively considers the trip route for being suitable for user that the factors such as condition of road surface, path length are planned；
Path adjusts module M500, for judging in first planning path with the presence or absence of the traveling with the thunder cloud
The distance in path is less than or equal to the location point of pre-determined distance, if it is, adjustment first planning path, so that described first
Each location point is all larger than pre-determined distance at a distance from the travel path of the thunder cloud in planning path.As shown in Fig. 2, being one
The schematic diagram of path adjustment in specific example.Wherein, the thunder cloud detected is on the position of L1, the prediction traveling road of thunder cloud
Diameter is L2.When for the stroke planning path of departure place A to destination B, the first planning path is the path ADCB.However, wherein C
Point position is less than pre-determined distance with thunder cloud travel path L2 distance d, therefore readjusts the first planning path, obtains new
Planning path ADEB, so that the distance between travel path L2 of each location point and thunder cloud is equal in the first planning path
Greater than pre-determined distance.
In this embodiment, the intelligent path planning system based on weather cloud atlas further includes identification model training module
M600, for training thunder cloud identification model, the thunder cloud identification model is using cloud atlas area image training set to being based on
What the deep learning model of convolutional neural networks was trained, which includes multiple in the cloud atlas area image training set, includes
Have thunder cloud cloud atlas area image and it is multiple do not include thunder cloud cloud atlas area image, and in each cloud atlas area image into
Whether row has the label of thunder cloud, and the input of the thunder cloud identification model is the cloud atlas area image of pre-set dimension, exports and is
There are the probability of thunder cloud in the cloud atlas region.The deep learning model based on convolutional neural networks may include convolution
Layer, pond layer, full articulamentum and softmax classification layer, the convolutional layer are connected by the pond layer with the full articulamentum
It connects, the output of the full articulamentum inputs the softmax classification layer, highest from softmax classification layer acquisition probability
The option sequence of menus at different levels, the output as the convolutional neural networks model.
Wherein, the function of convolutional layer is that feature extraction is carried out to input data, and internal includes multiple convolution kernels, and group is coiled
Each element of product core corresponds to a weight coefficient and a departure (bias vector), is similar to a feed forward neural
The neuron (neuron) of network.Multiple neurons in the region being closely located in each neuron and preceding layer in convolutional layer
It is connected, the size in region depends on the size of convolution kernel.After convolutional layer carries out feature extraction, the characteristic pattern of output can be passed
Feature selecting and information filtering are carried out to pond layer.Pond layer includes presetting pond function, and function is will be in characteristic pattern
The result of a single point replaces with the characteristic pattern statistic of its adjacent area.Pond layer choosing takes pond region and convolution kernel scanning feature
Figure step is identical, You Chihua size, step-length and filling control.Full articulamentum in convolutional neural networks is equivalent to conventional feed forward mind
Through the hidden layer in network.Full articulamentum is located at the decline of convolutional neural networks hidden layer, and only to other full articulamentums
Transmit signal.Characteristic pattern can lose Space expanding in full articulamentum, be expanded as vector and pass through excitation function.Finally
The output valve of one layer of full articulamentum is delivered to an output, can use softmax logistic regression (softmax
Regression) classify, which is alternatively referred to as softmax classification layer.
The thunder cloud identification module M100 is identified using following steps whether there is thunder cloud in the weather cloud atlas:
The weather cloud atlas is divided into the cloud atlas region of multiple pre-set dimensions by the thunder cloud identification module M100, will be more
A cloud atlas region sequentially inputs trained thunder cloud identification model, if there are the probability of thunder cloud in a cloud atlas region
Greater than predetermined probabilities threshold value, then judge that there are thunder clouds in the cloud atlas region；
The thunder cloud identification module M100 judges to deposit in a weather cloud atlas with the presence or absence of at least one described cloud atlas region
In thunder cloud, if it is, determining that there are thunder clouds in the weather cloud atlas.
In this embodiment, the determination thunder cloud identification module M100 determines that there are thunderstorms in the weather cloud atlas
Further include following steps after cloud:
There are each cloud atlas regions of the thunder cloud with the presence or absence of adjacent for the thunder cloud identification module M100 judgement
Cloud atlas region, if there is there are the thunder clouds in adjacent cloud atlas region, then by the thunder cloud in adjacent cloud atlas region
It is recorded as the same thunder cloud, and is numbered for each thunder cloud, the number of the thunder cloud in adjacent cloud atlas region is same
One.
The present invention using deep learning convolutional neural networks model automatic identification weather cloud atlas in thunder cloud, and
When preparing training set, the cloud atlas area image of the thunder cloud under different time different background, convenient accurate crawl thunder can be prepared
The feature of sudden and violent cloud, improves the accuracy rate of thunder cloud identification.And automatic identification is carried out by using convolutional neural networks model, is mentioned
The high efficiency of thunder cloud identification.
In this embodiment, the thunder cloud prediction module M200 calculates thunderstorm in the weather cloud atlas using following steps
The movement velocity and the direction of motion of cloud:
The thunder cloud prediction module M200 obtains multiple weather cloud atlas in the second preset time period in the past, obtains each
Position of the corresponding thunder cloud of number in every weather cloud atlas；Second preset time period can be a shorter time
Section, such as 2 hours, 5 hours etc., it specifically can according to need and selected and adjusted；
According to the position of acquisition time corresponding to every weather cloud atlas and each thunder cloud in every weather cloud atlas, in advance
Survey the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas.It specifically, can be according to each thunder cloud in each day
Historical position in gas cloud figure, carries out curve fitting, and predicts the travel path of thunder cloud, and become according to the variation of movement velocity
Gesture, predicted motion speed, and thunder cloud is calculated according to the distance between movement velocity and each location point and reaches each location point
Prediction arrival time.
In this embodiment, path adjustment module M500 be also used to judge the driving plan of travel departure place and
Whether at least one is less than or equal to the location point of pre-determined distance at a distance from the travel path of the thunder cloud in destination；Such as
Fruit is then merely to cannot achieve by adjusting planning path to avoid the travel path of thunder cloud, needs to carry out into one
The processing of step.
Specifically, if only departure place is less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud,
It determines the arrival time in the travel path of the thunder cloud with the departure place apart from nearest point, calculates arriving for nearest point
The wind that sets out is determined according to the mapping relations of preset time difference and risk class up to the time difference of time and departure time
The risk class that sets out is sent to user by dangerous grade, in the mapping relations of preset time difference and risk class, the time
Difference is smaller, and risk class is higher, conversely, time difference is bigger, risk class is lower；User can be according to the risk class that sets out
Risk assessment is carried out, in the case where risk class is relatively high, it is proposed that selection is gone on a journey or cancel plan of travel some other day；
If being only purposefully less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, it is determined that described
Arrival time in the travel path of thunder cloud with the destination apart from nearest point, calculate arrival time of nearest point with
The time difference of arrival time determines arrival risk class according to the mapping relations of preset time difference and risk class, will
The arrival risk class is sent to user；User can carry out risk assessment according to risk class is reached, in risk class
In the case where relatively high, it is proposed that plan of travel is cancelled in selection trip on some other day, change destination；
If departure place and destination are less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, divide
Not Ji Suan arrival time and departure time in the travel path of thunder cloud with departure place apart from nearest point first time it is poor
The second time difference of arrival time and arrival time in the travel path of value and thunder cloud with destination apart from nearest point,
It is determined respectively according to first time difference and the second time difference and sets out risk class and reach risk class, and be sent to use
Family.Whether user can need to change plan of travel according to set out risk class and arrival risk class comprehensive assessment, facilitate use
Possible risk is predicted at family in advance, and risk is hidden in selection as early as possible.
In this embodiment, the path adjusts module M500 for risk class and/or the arrival risk class of setting out
It is sent to after user, is also used to receive the driving plan of travel modification information of user, is repaired according to the driving plan of travel of user
Convert to the routing of breath modification user.After having modified the routing of user, can further it judge again modified
Whether routing will receive the influence of the travel path of thunder cloud.
As shown in figure 3, the embodiment of the present invention also provides a kind of intelligent paths planning method based on weather cloud atlas, using institute
The intelligent path planning system based on weather cloud atlas stated, described method includes following steps:
Weather cloud atlas is obtained, is identified in the weather cloud atlas with the presence or absence of thunder cloud；
If there is thunder cloud, then the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas are calculated；
The institute in the first preset time period is predicted according to the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas
State the travel path of thunder cloud；
The routing in the first preset time period of user is obtained, judging, which whether there is in the routing, driving
Plan of travel；
If there is driving plan of travel, then obtains the plan travel time corresponding to the driving plan of travel, sets out
Ground and destination；
Driving path planning is carried out according to the driving plan of travel, obtains the first planning path；
Judge pre- with the presence or absence of being less than or equal at a distance from the travel path of the thunder cloud in first planning path
If the location point of distance；
If it is, adjustment first planning path so that in first planning path each location point with it is described
The distance of the travel path of thunder cloud is all larger than pre-determined distance.
In this embodiment, the intelligent paths planning method based on weather cloud atlas further includes trained thunder cloud identification mould
The step of type, the trained thunder cloud identification model include the following steps:
Acquire it is multiple include thunder cloud cloud atlas area image and it is multiple do not include thunder cloud cloud atlas area image, and
It is made whether the label of thunder cloud in each cloud atlas area image, cloud atlas administrative division map is added in the cloud atlas area image after label
As training set；
The deep learning model based on convolutional neural networks is trained using the cloud atlas area image training set
To shown thunder cloud identification model, the input of the thunder cloud identification model is the cloud atlas area image of pre-set dimension, exports and is
There are the probability of thunder cloud in the cloud atlas region；
As shown in figure 4, whether there is thunder cloud in the identification weather cloud atlas, include the following steps:
The weather cloud atlas is divided into the cloud atlas region of multiple pre-set dimensions by the thunder cloud identification module, by multiple institutes
It states cloud atlas region and sequentially inputs trained thunder cloud identification model, if there are the probability of thunder cloud to be greater than in a cloud atlas region
Predetermined probabilities threshold value then judges that there are thunder clouds in the cloud atlas region；
The thunder cloud identification module judges that there are thunders with the presence or absence of at least one described cloud atlas region in a weather cloud atlas
Sudden and violent cloud, if it is, determining that there are thunder clouds in the weather cloud atlas.
In intelligent paths planning method of this kind based on weather cloud atlas, the specific implementation of each step can be adopted
It is not superfluous herein with the function implementation of each functional module in the above-mentioned intelligent path planning system based on weather cloud atlas
It states.
Compared with prior art, using the intelligent path planning system and method based on weather cloud atlas in the invention,
It has the following beneficial effects:
The present invention is based on the travel paths of weather cloud atlas look-ahead thunder cloud, help user to pacify in advance according to the stroke of user
Row avoids the trip route of thunder cloud, is conducive to user's safety；The present invention is based on the convolutional neural networks moulds of deep learning
Thunder cloud in type automatic identification weather cloud atlas, and according to the traveling of the dynamic prediction thunder cloud of thunder cloud in multiple weather cloud atlas
Direction and travel speed, so that thunder cloud travel path quick predict is realized, to improve processing speed；Further, this hair
It is bright that plan of travel is directly got from the routing of user, the path for avoiding insecurity factor is arranged for user in advance, is mentioned
High user's safety, and facilitate user to make strain early in the case where being likely encountered bad weather and arrange.
In this description, the present invention is described with reference to its specific embodiment.But it is clear that can still make
Various modifications and alterations are without departing from the spirit and scope of the invention.Therefore, the description and the appended drawings should be considered as illustrative
And not restrictive.
Claims (8)
1. a kind of intelligent path planning system based on weather cloud atlas, which is characterized in that the system comprises:
Thunder cloud identification module is analyzed in the weather cloud atlas for obtaining weather cloud atlas with the presence or absence of thunder cloud；
Thunder cloud prediction module when for recognizing thunder cloud in the weather cloud atlas, calculates thunderstorm in the weather cloud atlas
The movement velocity and the direction of motion of cloud, and existed according to movement velocity and the direction of motion prediction of thunder cloud in the weather cloud atlas
The travel path of the thunder cloud in first preset time period；
Stroke obtains module, and the routing in the first preset time period for obtaining user judges in the routing
With the presence or absence of there is driving plan of travel；
Path planning module, for there are when driving plan of travel, obtain the driving trip meter in the routing of user
Corresponding plan departure time, departure place, destination and arrival time are drawn, and is opened according to the driving plan of travel
Vehicle path planning obtains the first planning path；
Path adjusts module, for judge in first planning path with the presence or absence of with the travel path of the thunder cloud away from
From the location point for being less than or equal to pre-determined distance, if it is, first planning path is adjusted, so that first planning path
In each location point be all larger than pre-determined distance at a distance from the travel path of the thunder cloud.
2. the intelligent path planning system according to claim 1 based on weather cloud atlas, which is characterized in that the system is also
Including identification model training module, for training thunder cloud identification model, the thunder cloud identification model is using cloud atlas region
Training set of images is trained the deep learning model based on convolutional neural networks, the cloud atlas area image training
The cloud atlas area image for thunder cloud of concentrating that include multiple include and it is multiple do not include thunder cloud cloud atlas area image, and it is each
The label of thunder cloud is made whether in cloud atlas area image, the input of the thunder cloud identification model is the cloud atlas of pre-set dimension
Area image exports as there are the probability of thunder cloud in the cloud atlas region；
The thunder cloud identification module is identified using following steps whether there is thunder cloud in the weather cloud atlas:
The weather cloud atlas is divided into the cloud atlas region of multiple pre-set dimensions by the thunder cloud identification module, by multiple clouds
Graph region sequentially inputs trained thunder cloud identification model, presets if be greater than in a cloud atlas region there are the probability of thunder cloud
Probability threshold value then judges that there are thunder clouds in the cloud atlas region；
The thunder cloud identification module judges in a weather cloud atlas with the presence or absence of at least one described cloud atlas region there are thunder cloud,
If it is, determining that there are thunder clouds in the weather cloud atlas.
3. the intelligent path planning system according to claim 2 based on weather cloud atlas, which is characterized in that the determining institute
It states thunder cloud identification module to determine in the weather cloud atlas there are after thunder cloud, further includes following steps:
There are each cloud atlas regions of the thunder cloud whether there is adjacent cloud atlas region for the thunder cloud identification module judgement,
If there is there are the thunder clouds in adjacent cloud atlas region, then the thunder cloud in adjacent cloud atlas region is recorded as same
A thunder cloud, and be numbered for each thunder cloud, the number of the thunder cloud in adjacent cloud atlas region is same.
4. the intelligent path planning system according to claim 3 based on weather cloud atlas, which is characterized in that the thunder cloud
Prediction module calculates the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas using following steps:
The thunder cloud prediction module obtains multiple weather cloud atlas in the second preset time period in the past, and it is right to obtain each number institute
Position of the thunder cloud answered in every weather cloud atlas；
According to the position of acquisition time corresponding to every weather cloud atlas and each thunder cloud in every weather cloud atlas, institute is predicted
State the movement velocity and the direction of motion of thunder cloud in weather cloud atlas.
5. the intelligent path planning system according to claim 1 based on weather cloud atlas, which is characterized in that the path tune
Mould preparation block be also used to judge in the departure place and destination of the driving plan of travel whether at least one and the thunder cloud
Travel path distance be less than or equal to pre-determined distance location point；
If only departure place is less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, it is determined that the thunderstorm
It arrival time in the travel path of cloud with the departure place apart from nearest point, calculates arrival time of nearest point and sets out
The time difference of time determines the risk class that sets out according to the mapping relations of preset time difference and risk class, will be described
The risk class that sets out is sent to user；
If being only purposefully less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, it is determined that the thunderstorm
Arrival time in the travel path of cloud with the destination apart from nearest point calculates arrival time and the arrival of nearest point
The time difference of time determines arrival risk class according to the mapping relations of preset time difference and risk class, will be described
It reaches risk class and is sent to user；
If departure place and destination are less than or equal to pre-determined distance at a distance from the travel path of the thunder cloud, count respectively
Calculate in the travel path of thunder cloud with departure place apart from the nearest arrival time of point and the first time difference of departure time and
The second time difference of arrival time and arrival time in the travel path of thunder cloud with destination apart from nearest point, respectively
It is determined according to first time difference and the second time difference and sets out risk class and reach risk class, and be sent to user.
6. the intelligent path planning system according to claim 5 based on weather cloud atlas, which is characterized in that the path tune
Mould preparation block risk class and/or reaches described set out after risk class is sent to user, is also used to receive the driving of user
Plan of travel modification information modifies the routing of user according to the driving plan of travel modification information of user.
7. a kind of intelligent paths planning method based on weather cloud atlas, which is characterized in that use any one of claims 1 to 6
The intelligent path planning system based on weather cloud atlas, described method includes following steps:
Weather cloud atlas is obtained, is identified in the weather cloud atlas with the presence or absence of thunder cloud；
If there is thunder cloud, then the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas are calculated；
The thunder in the first preset time period is predicted according to the movement velocity and the direction of motion of thunder cloud in the weather cloud atlas
The travel path of sudden and violent cloud；
The routing in the first preset time period of user is obtained, judging, which whether there is in the routing, has driving to go on a journey
Plan；
If there is driving plan of travel, then obtain the plan travel time corresponding to the driving plan of travel, departure place and
Destination；
Driving path planning is carried out according to the driving plan of travel, obtains the first planning path；
Judge in first planning path with the presence or absence of be less than or equal at a distance from the travel path of the thunder cloud it is default away from
From location point；
If it is, adjustment first planning path, so that each location point and the thunderstorm in first planning path
The distance of the travel path of cloud is all larger than pre-determined distance.
8. the intelligent paths planning method according to claim 7 based on weather cloud atlas, which is characterized in that the method is also
Include the steps that training thunder cloud identification model, the trained thunder cloud identification model include the following steps:
Acquire it is multiple include thunder cloud cloud atlas area image and it is multiple do not include thunder cloud cloud atlas area image, and it is each
It is made whether the label of thunder cloud in cloud atlas area image, cloud atlas area image instruction is added in the cloud atlas area image after label
Practice collection；
The deep learning model based on convolutional neural networks is trained to obtain institute using the cloud atlas area image training set
Show thunder cloud identification model, the input of the thunder cloud identification model is the cloud atlas area image of pre-set dimension, is exported as the cloud
There are the probability of thunder cloud in graph region；
It whether there is thunder cloud in the identification weather cloud atlas, include the following steps:
The weather cloud atlas is divided into the cloud atlas region of multiple pre-set dimensions by the thunder cloud identification module, by multiple clouds
Graph region sequentially inputs trained thunder cloud identification model, presets if be greater than in a cloud atlas region there are the probability of thunder cloud
Probability threshold value then judges that there are thunder clouds in the cloud atlas region；
The thunder cloud identification module judges in a weather cloud atlas with the presence or absence of at least one described cloud atlas region there are thunder cloud,
If it is, determining that there are thunder clouds in the weather cloud atlas.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910808259.1A CN110487294A (en) | 2019-08-29 | 2019-08-29 | Intelligent path planning system and method based on weather cloud atlas |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910808259.1A CN110487294A (en) | 2019-08-29 | 2019-08-29 | Intelligent path planning system and method based on weather cloud atlas |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110487294A true CN110487294A (en) | 2019-11-22 |
Family
ID=68555116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910808259.1A Pending CN110487294A (en) | 2019-08-29 | 2019-08-29 | Intelligent path planning system and method based on weather cloud atlas |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110487294A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT202000004774A1 (en) * | 2020-03-06 | 2021-09-06 | Techno Sky S R L | Airport weather station |
| WO2022001545A1 (en) * | 2020-06-28 | 2022-01-06 | 中兴通讯股份有限公司 | Route planning method and device and computer-readable storage medium |
| CN114234997A (en) * | 2021-12-28 | 2022-03-25 | 佛山沐朝科技有限公司 | Path planning method, device, electronic device and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101963960A (en) * | 2009-07-22 | 2011-02-02 | 刘旸 | Travel route and scheduling generation method and server |
| CN103438894A (en) * | 2013-08-02 | 2013-12-11 | 浙江吉利汽车研究院有限公司 | Automobile-mounted navigation system and method |
| CN104200281A (en) * | 2014-08-25 | 2014-12-10 | 清华大学 | Method and system for predicting thunder cloud moving path based on lightning location system |
| CN105323301A (en) * | 2014-08-01 | 2016-02-10 | 宏达国际电子股份有限公司 | Mobile device with weather forecast |
| CN105466440A (en) * | 2015-11-18 | 2016-04-06 | 爱国者电子科技有限公司 | Navigation device for optimizing routes by utilization of weather forecast information, navigation system and method |
| CN106023177A (en) * | 2016-05-14 | 2016-10-12 | 吉林大学 | Thunderstorm cloud cluster identification method and system for meteorological satellite cloud picture |
| CN106887055A (en) * | 2017-01-23 | 2017-06-23 | 广州博进信息技术有限公司 | Flight is jolted method for early warning and its system |
| CN108919252A (en) * | 2018-04-03 | 2018-11-30 | 河北泽华伟业科技股份有限公司 | Stormy weather automatic tracing navigation system |
| CN109325960A (en) * | 2018-11-20 | 2019-02-12 | 南京信息工程大学 | A kind of infrared cloud image cyclone analysis method and analysis system |
| CN110044373A (en) * | 2019-05-27 | 2019-07-23 | 北京气象在线科技有限公司 | A kind of fining traffic weather information on services generation method |
-
2019
- 2019-08-29 CN CN201910808259.1A patent/CN110487294A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101963960A (en) * | 2009-07-22 | 2011-02-02 | 刘旸 | Travel route and scheduling generation method and server |
| CN103438894A (en) * | 2013-08-02 | 2013-12-11 | 浙江吉利汽车研究院有限公司 | Automobile-mounted navigation system and method |
| CN105323301A (en) * | 2014-08-01 | 2016-02-10 | 宏达国际电子股份有限公司 | Mobile device with weather forecast |
| CN104200281A (en) * | 2014-08-25 | 2014-12-10 | 清华大学 | Method and system for predicting thunder cloud moving path based on lightning location system |
| CN105466440A (en) * | 2015-11-18 | 2016-04-06 | 爱国者电子科技有限公司 | Navigation device for optimizing routes by utilization of weather forecast information, navigation system and method |
| CN106023177A (en) * | 2016-05-14 | 2016-10-12 | 吉林大学 | Thunderstorm cloud cluster identification method and system for meteorological satellite cloud picture |
| CN106887055A (en) * | 2017-01-23 | 2017-06-23 | 广州博进信息技术有限公司 | Flight is jolted method for early warning and its system |
| CN108919252A (en) * | 2018-04-03 | 2018-11-30 | 河北泽华伟业科技股份有限公司 | Stormy weather automatic tracing navigation system |
| CN109325960A (en) * | 2018-11-20 | 2019-02-12 | 南京信息工程大学 | A kind of infrared cloud image cyclone analysis method and analysis system |
| CN110044373A (en) * | 2019-05-27 | 2019-07-23 | 北京气象在线科技有限公司 | A kind of fining traffic weather information on services generation method |
Non-Patent Citations (2)
| Title |
|---|
| 周永章等: "《地球科学大数据挖掘与机器学习》", 30 September 2018, 中山大学出版社 * |
| 李雄: "飞行危险天气下的航班改航路径规划研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT202000004774A1 (en) * | 2020-03-06 | 2021-09-06 | Techno Sky S R L | Airport weather station |
| WO2022001545A1 (en) * | 2020-06-28 | 2022-01-06 | 中兴通讯股份有限公司 | Route planning method and device and computer-readable storage medium |
| CN114234997A (en) * | 2021-12-28 | 2022-03-25 | 佛山沐朝科技有限公司 | Path planning method, device, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109658695B (en) | Multi-factor short-term traffic flow prediction method | |
| CN110487294A (en) | Intelligent path planning system and method based on weather cloud atlas | |
| Thomas et al. | Predictions of urban volumes in single time series | |
| CN112700663A (en) | Multi-agent intelligent signal lamp road network control method based on deep reinforcement learning strategy | |
| CN110503104B (en) | Short-time remaining parking space quantity prediction method based on convolutional neural network | |
| CN110956807B (en) | Highway flow prediction method based on combination of multi-source data and sliding window | |
| CN114023062B (en) | Traffic flow information monitoring method based on deep learning and edge calculation | |
| CN112613225B (en) | Intersection traffic state prediction method based on neural network cell transmission model | |
| CN103839418B (en) | A kind of adaptive city expressway ring road kinetic-control system | |
| CN109143408B (en) | Dynamic region combined short-time rainfall forecasting method based on MLP | |
| CN113436433B (en) | Efficient urban traffic outlier detection method | |
| CN109191845A (en) | A kind of public transit vehicle arrival time prediction technique | |
| CN107195177A (en) | Based on Forecasting Methodology of the distributed memory Computational frame to city traffic road condition | |
| CN115223063A (en) | Unmanned aerial vehicle remote sensing wheat new variety lodging area extraction method and system based on deep learning | |
| CN110796317B (en) | Urban taxi scheduling method based on demand prediction | |
| CN114325879A (en) | Quantitative precipitation correction method based on classification probability | |
| CN112967493A (en) | Neural network-based prediction method for vehicle passing intersection travel time | |
| CN109800908A (en) | Signalized intersections operating status prediction technique and system based on LSTM model | |
| CN113674524A (en) | LSTM-GASVR-based multi-scale short-time traffic flow prediction modeling and prediction method and system | |
| CN112860782A (en) | Pure electric vehicle driving range estimation method based on big data analysis | |
| CN113449780A (en) | In-road berth occupancy prediction method based on random forest and LSTM neural network | |
| CN116797274A (en) | Shared bicycle demand prediction method based on Attention-LSTM-LightGBM | |
| CN112036598A (en) | Charging pile use information prediction method based on multi-information coupling | |
| CN117238126A (en) | Traffic accident risk assessment method under continuous flow road scene | |
| CN116050581A (en) | Smart city subway driving scheduling optimization method and Internet of things system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |
|
| RJ01 | Rejection of invention patent application after publication |