CN108692733A - A kind of map method for building up and map based on neural network establish system - Google Patents
A kind of map method for building up and map based on neural network establish system Download PDFInfo
- Publication number
- CN108692733A CN108692733A CN201810203065.4A CN201810203065A CN108692733A CN 108692733 A CN108692733 A CN 108692733A CN 201810203065 A CN201810203065 A CN 201810203065A CN 108692733 A CN108692733 A CN 108692733A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- map
- node
- data
- neural network
- 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
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
- G01C21/30—Map- or contour-matching
- G01C21/32—Structuring or formatting of map data
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The map method for building up and map that the invention discloses a kind of based on neural network establish system, are related to management map field.A kind of map method for building up based on neural network, including:Vehicle acquires the first map datum, and first map datum includes the first road condition data in the first preset range of current location and the current vehicle position of the vehicle;The vehicle obtains the node map data of the node from the main node in neural network;The vehicle establishs or updates the real-time map of the vehicle in conjunction with first map datum, node map data and static map.The present invention adjusts original plan route in time convenient for user, reduces influence of the accident for user so that user can quickly arrive at, and increase user goes out line efficiency.
Description
Technical field
The present invention relates to management map field, espespecially a kind of map method for building up and map based on neural network establish system
System.
Background technology
With the rapid development of industry, requirement of the people for clothing, food, lodging and transportion -- basic necessities of life is also being continuously improved.And the aspect being early expert at,
Automobile has come into huge numbers of families, and the appearance of automobile greatly facilitates the trip of people, for no matter for it is longer away from
With a distance from still short, people and object rapidly can be transported to target location by automobile, save the travel time of people,
The trip of people is no longer difficult.
Since Chinese topography is broad, road conditions are complicated, and for unfamiliar road conditions, people are still difficult to reach mesh
Cursor position.Therefore, the map system on automobile comes into being, and during people drive, the map system on vehicle also can
The road conditions near position and user's vehicle for enough reminding user current facilitate people and reach the destination that do not went,
Reduce the difficulty of people's trip.
Currently, the map system on automobile is merely able to provide a static map, or by navigation system by automobile
The current location direction of motion be labeled in map.But it is practical on the run, frequent occurrence traffic congestion, road construction or
The situations such as car accident, and cause vehicle that can not continue to travel with the route, user needs to adjust original plan route, causes
User needs to take more time and can arrive at, and reduce people goes out line efficiency.
Invention content
The map method for building up and map that the object of the present invention is to provide a kind of based on neural network establish system, convenient for using
Family adjusts original plan route in time, reduces influence of the accident for user so that user can quickly reach purpose
Ground, increase user goes out line efficiency.
Technical solution provided by the invention is as follows:
A kind of map method for building up based on neural network, including:S10, vehicle the first map datum of acquisition, described first
Map datum includes the first road conditions number in the first preset range of current location and the current vehicle position of the vehicle
According to;S20, the vehicle obtain the node map data of the node from the main node in neural network;Multiple vehicles are constituted
The neural network, each vehicle form the node of a neural network;The vehicle being connect with the vehicle direct communication
Be the vehicle main node, vehicle connect with the vehicle indirect communication for the vehicle secondary nodes;S30,
The vehicle establishs or updates the real-time of the vehicle in conjunction with first map datum, node map data and static map
Map;The vehicle obtains node map data of the data information as the vehicle from the real-time map;The data
Information include in the neural network in the position of each node and the neural network each node current location first
Road condition data.
Further, further comprising the steps of:The input operation of S40, vehicle parsing user, obtain rising for user's stroke
Point information and endpoint information;S41, the vehicle establish or update institute according to the real-time map, origin information and endpoint information
State the first guidance path of vehicle;S60, the vehicle navigate according to first guidance path.
Further, further comprising the steps of:S50, the vehicle obtain respectively from remaining all node in neural network
The node guidance path of a node;The node guidance path is the first guidance path that each node generates each self-generating;
S51, the vehicle combine the node guidance path of remaining all node, optimize the first guidance path of the vehicle.
Further, step S30 further includes:S31, the vehicle add first map datum on static map,
Complete establising or updating for the real-time map;On real-time map described in S32, the vehicle detection with the presence or absence of the node
The total data information of diagram data;S33, when in the vehicle detection to the real-time map be not present the node map data
Total data information when, the vehicle chosen from the node map data be not present on the real-time map data letter
Breath, and using the data information as data information to be added;S36, the vehicle are by the data information addition to be added in institute
It states on real-time map, redirects and execute step S32;S37, when there are node map numbers in the vehicle detection to the real-time map
According to total data information when, real-time map update is completed.
Further, further include after step S33:Whether data information to be added described in S34, the vehicle detection is from institute
State the first road condition data of the vehicle obtained in node map data;S35, when the vehicle detection to the number to be added
It is believed that when breath is the first road condition data of the vehicle obtained from the node map data, the vehicle waits adding by described
Add data information to be deleted from the node map data, and redirects and execute step S32;Otherwise, it redirects and executes step S36.
An object of the present invention also resides in a kind of map based on neural network of offer and establishes system, including nerve net
Network, the neural network are made of multiple vehicles, each described vehicle forms a neural network node;It is straight with the vehicle
The vehicle for connecing communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is the vehicle
Secondary nodes;Vehicle includes:Locating module, the current location for detecting vehicle, and the current location of the vehicle is sent
Give management map module;Sensor assembly, for detecting road conditions of the vehicle under current location in the first preset range, and
The first road condition data is formed, and first road condition data is sent to management map module;Wireless communication module is used for from god
Through the node map data for obtaining the node at remaining node in network;Management map module, the management map module into
One step includes:Map submodule, including static map is established in conjunction with the first map datum, node map data and static map
Or the real-time map of the update vehicle;First map datum includes that the current location of the vehicle, the vehicle are current
The first road condition data in the first preset range of position;Data extracting sub-module, for obtaining data from the real-time map
Node map data of the information as the vehicle;The data information include the position of each node in the neural network,
And in the neural network each node current location the first road condition data.
Further, the management map module further includes:User's analyzing sub-module, the input for parsing user operate,
Obtain the origin information and endpoint information of user's stroke;Path planning submodule, for according to the real-time map, origin information
And endpoint information, establish or update the first guidance path of the vehicle;The vehicle further includes:Vehicle control module, it is described
Vehicle control module is navigated according to first guidance path.
Further, the vehicle further includes:The wireless communication module obtains from remaining all node in neural network
Take the node guidance path of each node;The node guidance path is the first navigation road that each node generates each self-generating
Diameter;The path planning submodule combines the node guidance path of remaining all node, optimizes the first navigation road of the vehicle
Diameter.
Further, the management map module further includes:Data Detection submodule is on the real-time map for detecting
It is no that there are the total data information of the node map data;The map submodule is used to add first map datum
On the static map, establising or updating for the real-time map is completed;Described in being detected when the Data Detection submodule
When the total data information of the node map data being not present on real-time map, the map submodule is from the node map
The data information being not present on the real-time map is chosen in data, and using the data information as data information to be added, institute
Map is stated to add the data information to be added on the real-time map.
Further, the management map module further includes:Submodule is verified, is for detecting the data information to be added
It is no for the first road condition data of the vehicle obtained from the node map data;When the verification submodule detects institute
When stating the first road condition data that data information to be added is the vehicle obtained from the node map data, the map
Submodule deletes the data information to be added from the node map data;Otherwise, the map will be described to be added
Data information adds on the real-time map.
Compared with prior art, a kind of map method for building up and map based on neural network provided by the invention establish system
System has the advantages that:
1, multiple vehicles can be in communication with each other connection by wireless module, and multiple vehicles form neural network, Mei Geche
All as a node of neural network;Each vehicle can detect the road conditions around its own, form the first road condition data,
The road condition data that each vehicle can be detected is stored in vehicle;And due to each vehicle can each vehicle can will
Its road conditions detected is sent to the vehicle on remaining node by neural network, can also receive the vehicle on remaining node
Road conditions, and above process cycle executes, the road conditions that each vehicle can be detected constantly are sent out, in neural network
Vehicle on node can constantly receive the road conditions of remaining node transmission;When traffic congestion, road construction or vehicle thing occur for somewhere
Therefore when waiting accidents, by neural network it is also possible that the vehicle on remaining not nigh node constantly learns the event of handling affairs
Occur, adjust original plan route in time convenient for user, reduces user's arrival place where the accident occurred and just change road plan
May, reduce influence of the accident for user so that user can quickly arrive at, and increase the trip effect of user
Rate.
2, when real-time map is established, after vehicle gets the origin information and endpoint information of user's stroke, vehicle can
It generates or updates the first guidance path, while vehicle, during generating the first guidance path, vehicle can avoid road automatically
On accident, and choose the path of optimization as the first guidance path, vehicle can be carried out according to first guidance path
Navigation, rapidly arrives at, reduces influence of the accident on road to vehicle.Simultaneously in vehicle operation, i.e.,
When making that accident occurs on the first guidance path to be formed, vehicle can also be learnt in time, and plan the first guidance path again, then
Influence of the secondary accident reduced on road to vehicle.
3, guidance path can also be mutually shared between each node, vehicle can be according to the navigation road of the vehicle in node
Diameter, clearly recognizes the vehicle trend of each node in neural network, and realizes the traveling trend of prediction surrounding vehicles, just
The accident on the first guidance path there may be traffic congestion can be predicted in vehicle, vehicle can change the first navigation road in time
Diameter reduces influence of remaining vehicle to this vehicle, further increases the running efficiency of vehicle.
4, for the vehicle of startup, due to vehicle detection to the first map datum by its own, directly detection obtains, vehicle
Can directly by the first map datum addition real-time map is formed on static map;And the node map number to receiving
Be also required to judge on current real-time map with the presence or absence of the data information in node map data according to, vehicle, vehicle without
Already existing data information is operated again again, simplifies the more new technological process of real-time map.
5, node map data are received to obtain by vehicle, can not learn correctness, the data information on node map
When being clashed with the data information on real-time map, since the first map datum is obtained by vehicle itself detection, the first map
There is no the possibility of mistake, vehicles only to need using the first map datum as constant variable for data, then with changing remaining node
When data information on diagram data, you can obtain more accurate real-time map;And when this vehicle extracts updated real-time map
Node map data when, remaining vehicle can also obtain this vehicle collected current location the first map datum.
Description of the drawings
Below by a manner of clearly understandable, preferred embodiment is described with reference to the drawings, to a kind of based on neural network
Above-mentioned characteristic, technical characteristic, advantage and its realization method that map method for building up and map establish system are further described.
Fig. 1 is a kind of flow diagram of the map method for building up based on neural network of the present invention;
Fig. 2 is the flow diagram of another map method for building up based on neural network of the invention;
Fig. 3 is the flow diagram of another map method for building up based on neural network of the invention;
Fig. 4 is the structural schematic diagram that a kind of map based on neural network of the present invention establishes system;
Fig. 5 is the structural schematic diagram that a kind of map based on neural network of the present invention establishes neural network in system.
Drawing reference numeral explanation:10. central control module, 20. management map modules, 21. map submodules, 22. data carry
Submodule is taken, 23. user's analyzing sub-modules, 24. path planning submodules, 25. Data Detection submodules, 26. verify submodules,
27. map shows submodule, 30. locating modules, 40. sensor assemblies, 50. wireless communication modules, 60. vehicle control modules.
Specific implementation mode
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, control is illustrated below
The specific implementation mode of the present invention.It should be evident that drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, other are can also be obtained according to these attached drawings
Attached drawing, and obtain other embodiments.
To make simplified form, part related to the present invention is only schematically shown in each figure, they are not represented
Its practical structures as product.In addition, so that simplified form is easy to understand, there is identical structure or function in some figures
Component only symbolically depicts one of those, or has only marked one of those.Herein, "one" is not only indicated
" only this ", can also indicate the situation of " more than one ".
According to a kind of embodiment provided by the invention, as shown in Figure 1, a kind of map method for building up based on neural network,
Including:
S10, vehicle acquire the first map datum, and first map datum includes current location and the institute of the vehicle
State the first road condition data in the first preset range of current vehicle position.
S20, the vehicle obtain the node map data of the node from the main node in neural network.
Multiple vehicles constitute the neural network, each vehicle forms the node of a neural network;With it is described
The vehicle of vehicle direct communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is described
The secondary nodes of vehicle.
S30, the vehicle establish or update institute in conjunction with first map datum, node map data and static map
State the real-time map of vehicle.
The vehicle obtains node map data of the data information as the vehicle from the real-time map;The number
It is believed that breath include in the neural network in the position of each node and the neural network each node current location the
One road condition data.In the present embodiment, multiple vehicles can be in communication with each other connection by wireless module, and multiple vehicles form god
Through network, a node of each vehicle as neural network;Each vehicle can detect the road conditions around its own, be formed
First road condition data, the road condition data that each vehicle can be detected are stored in vehicle.
And due to each vehicle can the road conditions that can be detected of each vehicle remaining is sent to by neural network
Vehicle on node can also receive the road conditions of the vehicle on remaining node, and above process cycle executes, and each vehicle can
The road conditions detected are constantly sent out, and the vehicle on node in neural network can constantly receive the transmission of remaining node
Road conditions.
Specifically, each node of neural network can be communicated by wireless module, and wireless module refers mainly to WIFI module.
First preset range also refers mainly to the range for the road conditions that vehicle is able to detect that.It is also main logical when its current location of vehicle detection
It crosses GPS positioning system and positions its current location.
After vehicle receives the node map data of remaining node, vehicle can be according to current in node map data
The road conditions of the location determination position, the implementation map that vehicle is formed can show the road conditions under two positions;Above process speed
Comparatively fast, and between the individual nodes always on the one hand continuous service, real-time map can record this vehicle current location in real time
Road conditions, while vehicle also can quickly receive the road conditions of each node current location in neural network, and god is obtained convenient for user
Road conditions under coverage range through network.
More specifically, for example, when the accidents such as traffic congestion, road construction or car accident occur for somewhere, pass through nerve net
Network it is also possible that the vehicle on remaining not nigh node constantly learn this handle affairs thus generation, adjusted in time convenient for user
Original plan route reduces user and reaches place where the accident occurred and just changes the possibility of road plan, reduce accident for
The influence at family so that user can quickly arrive at, and increase user goes out line efficiency.
According to another embodiment provided by the invention, as shown in Fig. 2, a kind of map foundation side based on neural network
Method, including:
S10, vehicle acquire the first map datum, and first map datum includes current location and the institute of the vehicle
State the first road condition data in the first preset range of current vehicle position.
S20, the vehicle obtain the node map data of the node from the main node in neural network.
Multiple vehicles constitute the neural network, each vehicle forms the node of a neural network;With it is described
The vehicle of vehicle direct communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is described
The secondary nodes of vehicle.
S30, the vehicle establish or update institute in conjunction with first map datum, node map data and static map
State the real-time map of vehicle.
The vehicle obtains node map data of the data information as the vehicle from the real-time map;The number
It is believed that breath include in the neural network in the position of each node and the neural network each node current location the
One road condition data.
The input operation of S40, vehicle parsing user, obtain the origin information and endpoint information of user's stroke.
S41, the vehicle establish or update the vehicle according to the real-time map, origin information and endpoint information
First guidance path.
Optionally, S50, the vehicle obtain the node navigation of each node from remaining all node in neural network
Path;The node guidance path is the first guidance path that each node generates each self-generating.
Optionally, S51, the vehicle combine the node guidance path of remaining all node, optimize the first of the vehicle
Guidance path.
S60, the vehicle navigate according to first guidance path.
In the present embodiment, when real-time map foundation, after vehicle gets the origin information and endpoint information of user's stroke,
Vehicle can generate or update the first guidance path, while vehicle, during generating the first guidance path, vehicle can be automatic
The accident on road is avoided, and chooses the path of optimization as the first guidance path, vehicle can navigate according to described first
Path is navigated, and is rapidly arrived at, and influence of the accident on road to vehicle is reduced.
Simultaneously in vehicle operation, though if occurring vehicle when accident on the first guidance path formed can and
Shi get Zhi, and the first guidance path is planned again, again reduce influence of the accident on road to vehicle.
Guidance path can also be mutually shared between each node, vehicle can be according to the navigation road of the vehicle in node
Diameter, clearly recognizes the vehicle trend of each node in neural network, and realizes the traveling trend of prediction surrounding vehicles, just
The accident on the first guidance path there may be traffic congestion can be predicted in vehicle, vehicle can change the first navigation road in time
Diameter reduces influence of remaining vehicle to this vehicle, further increases the running efficiency of vehicle.
According to another embodiment provided by the invention, as shown in figure 3, a kind of map foundation side based on neural network
Method, including:
S10, vehicle acquire the first map datum, and first map datum includes current location and the institute of the vehicle
State the first road condition data in the first preset range of current vehicle position.
S20, the vehicle obtain the node map data of the node from the main node in neural network.
Multiple vehicles constitute the neural network, each vehicle forms the node of a neural network;With it is described
The vehicle of vehicle direct communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is described
The secondary nodes of vehicle.
S31, the vehicle add first map datum on static map, complete the foundation of the real-time map
Or update.
It whether there is the total data information of the node map data on real-time map described in S32, the vehicle detection.
S33, when in the vehicle detection to the real-time map be not present the node map data total data believe
When breath, the vehicle chooses the data information being not present on the real-time map from the node map data, and by the number
It is believed that breath is used as data information to be added.
Optionally, S34, whether data information to be added is to be obtained from the node map data described in the vehicle detection
First road condition data of the vehicle taken.
Optionally, S35, when the vehicle detection to the data information to be added is to obtain from the node map data
When the first road condition data of the vehicle taken, the vehicle is by the data information to be added from the node map data
It deletes, and redirects and execute step S32;Otherwise, it redirects and executes step S36.
The data information addition to be added on the real-time map, is redirected and executes step S32 by S36, the vehicle.
S37, when in the vehicle detection to the real-time map there are when the total data information of node map data, institute
Real-time map update is stated to complete.
The vehicle obtains node map data of the data information as the vehicle from the real-time map;The number
It is believed that breath include in the neural network in the position of each node and the neural network each node current location the
One road condition data.
In the present embodiment, for the vehicle of startup, since the first map datum that vehicle detection arrives directly is examined by its own
It measures, the addition of the first map datum directly can be formed real-time map by vehicle on static map;And the section to receiving
Point map datum, vehicle are also required to judge on current real-time map with the presence or absence of the data information in node map data, vehicle
Without again operating already existing data information again, the more new technological process of real-time map is simplified.
Node map data are received to obtain by vehicle, can not learn correctness, the data information on node map with
When data information on real-time map clashes, since the first map datum is obtained by vehicle itself detection, the first map number
According to there is no the possibility of mistake, vehicle is only needed using the first map datum as constant variable, then changes remaining node map
When data information in data, you can obtain more accurate real-time map;And when this vehicle extracts updated real-time map
When node map data, remaining vehicle can also obtain this vehicle collected current location the first map datum.
According to another embodiment provided by the invention, as shown in Figures 2 and 3, a kind of map based on neural network is built
Cube method, including:
S10, vehicle acquire the first map datum, and first map datum includes current location and the institute of the vehicle
State the first road condition data in the first preset range of current vehicle position.
S20, the vehicle obtain the node map data of the node from the main node in neural network.
Multiple vehicles constitute the neural network, each vehicle forms the node of a neural network;With it is described
The vehicle of vehicle direct communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is described
The secondary nodes of vehicle.
S31, the vehicle add first map datum on static map, complete the foundation of the real-time map
Or update.
It whether there is the total data information of the node map data on real-time map described in S32, the vehicle detection.
S33, when in the vehicle detection to the real-time map be not present the node map data total data believe
When breath, the vehicle chooses the data information being not present on the real-time map from the node map data, and by the number
It is believed that breath is used as data information to be added.
Optionally, S34, whether data information to be added is to be obtained from the node map data described in the vehicle detection
First road condition data of the vehicle taken.
Optionally, S35, when the vehicle detection to the data information to be added is to obtain from the node map data
When the first road condition data of the vehicle taken, the vehicle is by the data information to be added from the node map data
It deletes, and redirects and execute step S32;Otherwise, it redirects and executes step S36.
The data information addition to be added on the real-time map, is redirected and executes step S32 by S36, the vehicle.
S37, when in the vehicle detection to the real-time map there are when the total data information of node map data, institute
Real-time map update is stated to complete.
The vehicle obtains node map data of the data information as the vehicle from the real-time map;The number
It is believed that breath include in the neural network in the position of each node and the neural network each node current location the
One road condition data.
The input operation of S40, vehicle parsing user, obtain the origin information and endpoint information of user's stroke.
S41, the vehicle establish or update the vehicle according to the real-time map, origin information and endpoint information
First guidance path.
Optionally, S50, the vehicle obtain the node navigation of each node from remaining all node in neural network
Path;The node guidance path is the first guidance path that each node generates each self-generating.
Optionally, S51, the vehicle combine the node guidance path of remaining all node, optimize the first of the vehicle
Guidance path.
S60, the vehicle navigate according to first guidance path.
In the present embodiment, vehicle can be realized according to the real-time map, origin information and endpoint information, be established or more
First guidance path of the new vehicle, and combine the node guidance path of remaining all node optimizes the of the vehicle
One guidance path realizes update and optimization for the first guidance path.Vehicle can also be realized in conjunction with first map
Data, node map data and static map, establish or update the real-time map of the vehicle, get under neural network covering
Real-time map.Vehicle can rapidly obtain the running method for reaching destination locations, increase the running efficiency of vehicle.
According to a kind of embodiment provided by the invention, as shown in figure 4, a kind of map based on neural network establishes system,
Including neural network, the neural network is made of multiple vehicles, each described vehicle forms a neural network node;With
The vehicle of the vehicle direct communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is
The secondary nodes of the vehicle.
Vehicle includes:
Locating module 30, the current location for detecting vehicle, and the current location of the vehicle is sent to map pipe
Manage module 20.
Sensor assembly 40 for detecting road conditions of the vehicle under current location in the first preset range, and is formed
First road condition data, and first road condition data is sent to management map module 20.
Wireless communication module 50, the node map data for obtaining the node from remaining node in neural network;
The wireless communication module 50 obtains the node guidance path of each node from remaining all node in neural network.
Vehicle control module 60, the vehicle control module 60 are navigated according to first guidance path.
Management map module 20, the management map module 20 further comprise:
Map submodule 21, including static map is built in conjunction with the first map datum, node map data and static map
Real-time map that is vertical or updating the vehicle;The map submodule 21 is used to add first map datum described quiet
On state map, establising or updating for the real-time map is completed;First map datum include the vehicle current location,
The first road condition data in the first preset range of the current vehicle position.
Data extracting sub-module 22, data extracting sub-module, for obtaining data information conduct from the real-time map
The node map data of the vehicle;The data information includes the position of each node and described in the neural network
First road condition data of each node current location in neural network.
User's analyzing sub-module 23, the input for parsing user operate, and obtain the origin information and terminal of user's stroke
Information.
Path planning submodule 24, for according to the real-time map, origin information and endpoint information, establising or updating institute
State the first guidance path of vehicle;The path planning submodule 24 combines the node guidance path of remaining all node, optimization
First guidance path of the vehicle.
Data Detection submodule 25, for detecting the whole that whether there is the node map data on the real-time map
Data information;When the Data Detection submodule 25 detects that there is no the complete of the node map data on the real-time map
When portion's data information, the map submodule 21 chooses the number being not present on the real-time map from the node map data
It is believed that breath, and using the data information as data information to be added, the map is by the data information addition to be added in institute
It states on real-time map.
Submodule 26 is verified, for detecting whether the data information to be added is to be obtained from the node map data
The vehicle the first road condition data;
When the verification submodule 26 detects that the data information to be added is to be obtained from the node map data
The vehicle the first road condition data when, the map submodule 21 is by the data information to be added from the node map
It is deleted in data;Otherwise, the map adds the data information to be added on the real-time map.
Map shows submodule 27, for showing the real-time map, the first guidance path and remaining node navigation road
Diameter.
In the present embodiment, map establishes system in addition to above-mentioned module, further includes central control module 10, the center control
Molding block 10 is mainly cpu chip.
As shown in figure 5, including multiple vehicles in neural network, since multiple vehicles are led to by short-distance radio module
Letter, therefore, each vehicle is merely able to be communicated with remaining a certain range of vehicle.Using No. 1 vehicle as initial vehicle, and it is every
When the WIFI signal of a vehicle is merely able to be communicated with the vehicle of a distance, then No. 1 vehicle be capable of around all No. 2 vehicles into
Row communication, therefore, first nodes of No. 2 vehicles as No. 1 vehicle;But No. 3 vehicles can not be with No. 1 vehicle direct communication, and is merely able to lead to
No. 2 vehicles are crossed to be communicated with No. 1 vehicle, therefore, two-level node of No. 3 vehicles as No. 1 vehicle.It is formed and is made of multiple vehicles therefrom
Neural network.
In the present embodiment, the first nodes in same neural network are the main node of vehicle, same neural network
In remaining node, that is, vehicle secondary nodes.
Using No. 1 vehicle as initial vehicle, and when each vehicle can be communicated with the vehicle of two distances, No. 2 vehicles
Can be as the first nodes of No. 1 vehicle with No. 3 vehicles, and No. 4 vehicles can be as 2 grades of nodes of No. 1 vehicle;Due to No. 5 vehicles
There are 4 distances between No. 1 vehicle, when, there are when No. 3 vehicles, No. 5 vehicles can pass through No. 3 vehicles and No. 1 between No. 5 vehicles and No. 1 vehicle
Vehicle is communicated, and No. 5 vehicles can be as 3 grades of nodes of No. 1 vehicle;And when No. 3 vehicles are not present between No. 5 vehicles and No. 1 vehicle, only deposit
In No. 2 vehicles and No. 4 vehicles, when No. 5 vehicles are merely able to be communicated with No. 1 vehicle by No. 2 vehicles again by No. 4 vehicles, No. 5 vehicles can only
Three-level node enough as No. 1 vehicle;When No. 3 vehicles and No. 4 vehicles are not present between No. 5 vehicles and No. 1 vehicle, since No. 5 vehicles can not
It being communicated with No. 2 vehicles, therefore neural network only includes No. 1 vehicle and No. 2 vehicles, No. 5 vehicles belong in another neural network, until
Family Nujiang between No. 5 vehicles and No. 2 vehicles be less than two apart from when, No. 5 vehicles can belong to this neural network, and as No. 1 vehicle
Two-level node.
Specifically, as shown in Figure 4 and Figure 5, when vehicle just starts, central control module 10 can control map submodule
21 and sensor assembly 40 open.Central control module 10 refers mainly to cpu chip, and sensor assembly 40 includes mainly that image passes
Sensor, laser radar sensor, ultrasonic sensor and infrared sensor, sensor assembly 40 can include therein one
It is a or multiple.When central control module 10, which controls sensor assembly 40, to be opened, sensor assembly 40 can detect vehicle the
Load conditions in one preset range, the first preset range refer mainly to the range for the road conditions that vehicle is able to detect that.Meanwhile center
Control module 10 can control the work of locating module 30, obtain the current location of vehicle.Central control module 10 can receive
First map datum.
Later, since the vehicle foundation on node completes the establishment of real-time map, the center of the vehicle on remaining node
Control module 10 can also control the data on the extraction real-time map of data extracting sub-module 22, form node map data, it
The central control module 10 of rear vehicle can receive the node map data that the vehicle of neural network first nodes is sent, node
Include whole map datums that static map is removed on real-time map on map datum.
When central control module 10 receives the first map datum with node map data, central control module 10 can incite somebody to action
First map datum is sent to management map module 20 with node map data, the map submodule 21 in management map module 20
After receiving the first map datum, by the addition of the first map datum on static map, the foundation of real-time map is completed, and work as
Preceding real-time map can show the road conditions of the first preset range of current vehicle position.
Later, due to including the map datum in the first preset range of a part of current location in node map data,
Including the data information outside the first preset range of a part of current location;Since vehicle is initially opened, current location first is default
Remaining data information is not present outside range, real-time map can be whole by the data information outside the first preset range of current location
Addition is on real-time map.
Complete the primary update of real-time map, rear vehicle can begin setting up the first navigation in real time according to updated
Path, vehicle can start running according to the first guidance path of foundation, this system can be applied to full-automatic driving system or
Semi-automatic driving system.Map displaying submodule 27 refers mainly to display screen, and user can check current real-time on a display screen
Map and the current guidance path of vehicle.
Later, vehicle proceeds by second of update of real-time map, and update step is identical as above-mentioned update step, but
When management map module 20 receives the first map datum with node map data, due to included in the real-time map of vehicle
The data information of remaining a large amount of position, therefore, Data Detection submodule 25 needs to first detect whether node map data
It is upper to whether there is new data information, only when there are when new data information, and the data information is also in node map data
In the first preset range of current vehicle position model cannot be preset since vehicle can be consistently detected its current location first
Road conditions in enclosing illustrate that first nodes vehicle is sent when the data information is in the first preset range of current vehicle position
Node map data on there are mistakes;Only when the data information is not in the first preset range of current vehicle position, vehicle
It can update real-time map.
The renewal process of above-mentioned real-time map occurs in real time, and the renewal process of real-time map can be happened at adjacent node
Between, the real-time map in neural network can update rapidly in a relatively short period of time, according to pre- in current vehicle position first
If the priority in range is higher, the lower sequential update map datum of remaining range priority, each node in neural network
Updated map can be got always.Therefore, even if accident occurs for the somewhere under neural network covering, in neural network
Each node can receive the generation of the accident, the vehicle convenient for each node in neural network can avoid the thing
Therefore.
Specifically, it include No. 4 in the real-time map of No. 4 vehicles for example, No. 4 vehicles can obtain the road conditions of No. 5 vehicle current locations
The road conditions of vehicle current location and No. 5 vehicle current locations, similarly, in the real-time map of No. 3 vehicles can include No. 3 vehicles, No. 4 vehicles with
And the road conditions of No. 5 vehicle current locations, until No. 1 vehicle obtains the road conditions of the current location of all vehicles;The reversed fortune of the above process
Row, No. 1 vehicle are communicated to connect with No. 2 vehicles, and No. 2 vehicles can also obtain the road conditions of all current vehicle positions, the above process persistently into
It goes, the vehicle in neural network can obtain the road conditions of all positions.
When establishing the first guidance path, vehicle first can according to real-time map and origin information input by user and
Endpoint information determines first guidance path, but when for closing on peak on and off duty, and vehicle is travelled along the first guidance path
There is the possibility of traffic congestion later, neural network can mutually share guidance path, vehicle can predict neural network system it is interior its
The drive route of remaining vehicle, it is thus understood that when somewhere may block up, vehicle can optimize the first guidance path, reduce
Vehicle encounters the possibility of traffic congestion.
Simultaneously when vehicle collision occurs for road, remaining vehicle can also be understood by neural network in time, reduce vehicle
The possibility of a chain of collision, increases the safety of vehicle operation.
It should be noted that above-described embodiment can be freely combined as needed.The above is only the preferred of the present invention
Embodiment, it is noted that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, several improvements and modifications can also be made, these improvements and modifications also should be regarded as protection scope of the present invention.
Claims (10)
1. a kind of map method for building up based on neural network, which is characterized in that including:
S10, vehicle acquire the first map datum, and first map datum includes the current location of the vehicle and the vehicle
The first road condition data in the first preset range of current location;
S20, the vehicle obtain the node map data of the node from the main node in neural network;
Multiple vehicles constitute the neural network, each vehicle forms the node of a neural network;With the vehicle
The vehicle of direct communication connection is the main node of the vehicle, and the vehicle being connect with the vehicle indirect communication is the vehicle
Secondary nodes;
S30, the vehicle establish or update the vehicle in conjunction with first map datum, node map data and static map
Real-time map;
The vehicle obtains node map data of the data information as the vehicle from the real-time map;The data letter
Breath includes the first via of each node current location in the position of each node and the neural network in the neural network
Condition data.
2. a kind of map method for building up based on neural network according to claim 1, which is characterized in that further include following
Step:
The input operation of S40, vehicle parsing user, obtain the origin information and endpoint information of user's stroke;
S41, the vehicle establish or update the first of the vehicle according to the real-time map, origin information and endpoint information
Guidance path;
S60, the vehicle navigate according to first guidance path.
3. a kind of map method for building up based on neural network according to claim 2, which is characterized in that further include following
Step:
S50, the vehicle obtain the node guidance path of each node from remaining all node in neural network;The section
Point guidance path is the first guidance path that each node generates each self-generating;
S51, the vehicle combine the node guidance path of remaining all node, optimize the first guidance path of the vehicle.
4. a kind of map method for building up based on neural network according to claim 1 or 2 or 3, which is characterized in that step
S30 further includes:
S31, the vehicle add first map datum on static map, complete the foundation or more of the real-time map
Newly;
It whether there is the total data information of the node map data on real-time map described in S32, the vehicle detection;
S33, when in the vehicle detection to the real-time map be not present the node map data total data information when,
The vehicle chooses the data information being not present on the real-time map from the node map data, and by the data information
As data information to be added;
The data information addition to be added on the real-time map, is redirected and executes step S32 by S36, the vehicle;
S37, when in the vehicle detection to the real-time map there are when the total data information of node map data, the reality
When map rejuvenation complete.
5. a kind of map method for building up based on neural network according to claim 4, which is characterized in that step S33 it
After further include:
Whether data information to be added described in S34, the vehicle detection is the vehicle obtained from the node map data
The first road condition data;
S35, when the vehicle detection to the data information to be added is the vehicle obtained from the node map data
The first road condition data when, the vehicle deletes the data information to be added from the node map data, and jumps
Turn to execute step S32;Otherwise, it redirects and executes step S36.
6. a kind of map based on neural network establishes system, it is characterised in that:Including neural network, the neural network is by more
A vehicle is constituted, each described vehicle forms a neural network node;The vehicle being connect with the vehicle direct communication is
The main node of the vehicle, the vehicle being connect with the vehicle indirect communication are the secondary nodes of the vehicle;
Vehicle includes:
Locating module, the current location for detecting vehicle, and the current location of the vehicle is sent to management map module;
Sensor assembly for detecting road conditions of the vehicle under current location in the first preset range, and forms the first via
Condition data, and first road condition data is sent to management map module;
Wireless communication module, the node map data for obtaining the node from remaining node in neural network;
Management map module, the management map module further comprise:
Map submodule in conjunction with the first map datum, node map data and static map, is established or more including static map
The real-time map of the new vehicle;First map datum includes the current location of the vehicle, the current vehicle position
The first road condition data in first preset range;Data extracting sub-module, for obtaining data information from the real-time map
Node map data as the vehicle;The data information include in the neural network position of each node and
First road condition data of each node current location in the neural network.
7. a kind of map based on neural network according to claim 6 establishes system, it is characterised in that:
The management map module further includes:
User's analyzing sub-module, the input for parsing user operate, and obtain the origin information and endpoint information of user's stroke;
Path planning submodule, for according to the real-time map, origin information and endpoint information, establising or updating the vehicle
The first guidance path;
The vehicle further includes:
Vehicle control module, the vehicle control module are navigated according to first guidance path.
8. a kind of map based on neural network according to claim 7 establishes system, it is characterised in that:
The wireless communication module obtains the node guidance path of each node from remaining all node in neural network;Institute
It is the first guidance path that each node generates each self-generating to state node guidance path;
The path planning submodule combines the node guidance path of remaining all node, optimizes the first navigation road of the vehicle
Diameter.
9. a kind of map based on neural network described according to claim 6 or 7 or 8 establishes system, which is characterized in that described
Management map module further includes:
Data Detection submodule is believed for detecting the total data on the real-time map with the presence or absence of the node map data
Breath;
The map submodule is used to add first map datum in the static map, completes the real-time map
Establish or update;
When the Data Detection submodule detects the total data on the real-time map there is no the node map data
When information, the map submodule chooses the data information being not present on the real-time map from the node map data,
And using the data information as data information to be added, the map by the data information addition to be added it is described in real time
On figure.
10. a kind of map based on neural network according to claim 9 establishes system, which is characterized in that the map
Management module further includes:
Verify submodule, for detect the data information to be added whether be obtained from the node map data described in
First road condition data of vehicle;
When the verification submodule detect the data information to be added be obtain from the node map data described in
When the first road condition data of vehicle, the map submodule deletes the data information to be added from the node map data
It removes;Otherwise, the map adds the data information to be added on the real-time map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810203065.4A CN108692733A (en) | 2018-03-13 | 2018-03-13 | A kind of map method for building up and map based on neural network establish system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810203065.4A CN108692733A (en) | 2018-03-13 | 2018-03-13 | A kind of map method for building up and map based on neural network establish system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108692733A true CN108692733A (en) | 2018-10-23 |
Family
ID=63844475
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810203065.4A Pending CN108692733A (en) | 2018-03-13 | 2018-03-13 | A kind of map method for building up and map based on neural network establish system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108692733A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852342A (en) * | 2019-09-26 | 2020-02-28 | 京东城市(北京)数字科技有限公司 | Road network data acquisition method, device, equipment and computer storage medium |
| CN112585656A (en) * | 2020-02-25 | 2021-03-30 | 华为技术有限公司 | Method and device for identifying special road conditions, electronic equipment and storage medium |
| US12005909B2 (en) | 2020-03-13 | 2024-06-11 | Ford Global Technologies, Llc | Vehicle roof assembly |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1321742A2 (en) * | 2001-12-18 | 2003-06-25 | ZF Lemförder Metallwaren AG | Method and device for generating and updating a road map and/or a road condition map |
| CN103017783A (en) * | 2012-12-05 | 2013-04-03 | 中兴通讯股份有限公司 | Navigation method and system, map data management cloud end and data updating method thereof |
| CN103136952A (en) * | 2011-12-02 | 2013-06-05 | 上海博泰悦臻电子设备制造有限公司 | Navigation method and navigation service center |
| CN103971529A (en) * | 2014-04-28 | 2014-08-06 | 中国科学院深圳先进技术研究院 | Method and system for acquiring and estimating road conditions based on vehicle-to-vehicle communication |
-
2018
- 2018-03-13 CN CN201810203065.4A patent/CN108692733A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1321742A2 (en) * | 2001-12-18 | 2003-06-25 | ZF Lemförder Metallwaren AG | Method and device for generating and updating a road map and/or a road condition map |
| CN103136952A (en) * | 2011-12-02 | 2013-06-05 | 上海博泰悦臻电子设备制造有限公司 | Navigation method and navigation service center |
| CN103017783A (en) * | 2012-12-05 | 2013-04-03 | 中兴通讯股份有限公司 | Navigation method and system, map data management cloud end and data updating method thereof |
| CN103971529A (en) * | 2014-04-28 | 2014-08-06 | 中国科学院深圳先进技术研究院 | Method and system for acquiring and estimating road conditions based on vehicle-to-vehicle communication |
Non-Patent Citations (1)
| Title |
|---|
| 罗忠等: "车联网发展的探讨 ", 《电子技术与软件工程》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852342A (en) * | 2019-09-26 | 2020-02-28 | 京东城市(北京)数字科技有限公司 | Road network data acquisition method, device, equipment and computer storage medium |
| CN110852342B (en) * | 2019-09-26 | 2020-11-24 | 京东城市(北京)数字科技有限公司 | Road network data acquisition method, device, equipment and computer storage medium |
| CN112585656A (en) * | 2020-02-25 | 2021-03-30 | 华为技术有限公司 | Method and device for identifying special road conditions, electronic equipment and storage medium |
| CN112585656B (en) * | 2020-02-25 | 2022-06-17 | 华为技术有限公司 | Method and device for identifying special road conditions, electronic equipment and storage medium |
| US12005909B2 (en) | 2020-03-13 | 2024-06-11 | Ford Global Technologies, Llc | Vehicle roof assembly |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108039053B (en) | A kind of intelligent network connection traffic system | |
| CN112461256B (en) | Path planning method and device | |
| US10514697B2 (en) | Vehicle remote assistance mode | |
| CN102564440B (en) | Communication system and movable terminal | |
| CN105644559B (en) | A kind of unmanned method and system based on virtual rail and Moving Block Technology in UMT | |
| CN108628206A (en) | Road construction detecting system and method | |
| JP7241642B2 (en) | platooning system | |
| CN110494868A (en) | Vehicle electronic controlling device | |
| CN109644144A (en) | Radio network optimization | |
| CN110239547A (en) | Controller of vehicle, control method for vehicle and storage medium | |
| US20200101974A1 (en) | Device and method for selecting optimal travel route based on driving situation | |
| CN108091149A (en) | The dispatching method and device of a kind of emergency vehicle | |
| US11662735B2 (en) | Method and system for updating a control model for automatic control of at least one mobile unit | |
| CN105973248B (en) | Navigation system | |
| CN101377421B (en) | Apparatus and method for planning path | |
| CN109697875A (en) | Plan the method and device of driving trace | |
| JP2015212863A (en) | Traffic signal control device, traffic signal control method, and computer program | |
| CN109615887A (en) | Wisdom traffic network system signal guidance method | |
| CN108692733A (en) | A kind of map method for building up and map based on neural network establish system | |
| CN106205161A (en) | traffic information transmission method and device | |
| CN105261242A (en) | Systems and methods of airport traffic control | |
| CN106781556A (en) | A kind of traffic lights duration determination methods suitable for vehicle-mounted short distance communication network | |
| US20190180615A1 (en) | System and method for reducing delays in road traffic | |
| CN206961119U (en) | A kind of distributed memory system of intelligent driving automobile operation system and high-precision map | |
| CN107272689A (en) | A kind of Transport route planning and monitoring system and its path planning and monitoring method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181023 |
|
| WD01 | Invention patent application deemed withdrawn after publication |