CN106525063A - Autonomous refueling method of autonomous car and intelligent car - Google Patents
Autonomous refueling method of autonomous car and intelligent car Download PDFInfo
- Publication number
- CN106525063A CN106525063A CN201710024566.1A CN201710024566A CN106525063A CN 106525063 A CN106525063 A CN 106525063A CN 201710024566 A CN201710024566 A CN 201710024566A CN 106525063 A CN106525063 A CN 106525063A
- Authority
- CN
- China
- Prior art keywords
- intelligent vehicle
- value table
- traveling
- strategy value
- state
- 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
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
Abstract
The invention discloses an autonomous refueling method of an autonomous car, and an intelligent car, belongs to the field of intelligent cars, and aims to realize automatic refueling during driving of the intelligent car. The automatic refueling method comprises the following steps: forming a driving strategy value table of the intelligent car through autonomous learning of the intelligent car by utilizing a learning algorithm, wherein the driving strategy value table is formed by grille values of multiple grilles formed by subdividing a driving zone; automatically planning a driving route between the intelligent car and a refueling position based on the driving strategy value table; automatically driving to the refueling position according to the driving route, and refueling. The autonomous refueling method disclosed by the invention is used for autonomous refueling of automobiles.
Description
Technical field
The present invention relates to intelligent driving field, the autonomous oiling method of more particularly to a kind of autonomous driving vehicle and intelligent vehicle.
Background technology
In nearly century more than one, the appearance of automobile instead of conventional traffic means of transportation so that the life of people is more just
It is prompt.In the last few years, developing rapidly with the development of science and technology, especially intelligence computation, the research of autonomous driving vehicle becomes each
The focus that cart enterprise focuses on.In the recent period, Mai Kenxi has issued " determining 12 cutting edge technologies of a future economy " report, in report
12 cutting edge technologies are inquired into the future economy, the influence degree of society, the respective Jing of 12 technologies in 2025 has been estimated in analysis
Ji and social effectiveness.Wherein autonomous driving vehicle technology comes the 6th, and its power of influence estimation in 2025 is：Economic benefit is every
Year about 0.2～1.9 trillion dollars, social benefit can retrieve 3～150,000 life every year.As autonomous driving vehicle technology has
Larger market prospect and economical, societal benefits, each cart enterprise look forward to development using autonomous driving vehicle technology as future car
Core Development Technology.
As technology progress, the function of autonomous driving vehicle are more and more comprehensive, the scope of application is also more and more wider.From most
In fastlink, first can only realize that automatic Pilot has switched to gradually city road and can also realize automatic Pilot.Automatic Pilot skill
The progress of art, the popularization of scope so that, in addition to highway driving, some other links is also important all the more for autonomous driving vehicle,
Such as automatic oiling etc., these links ensure, could realize real automatic Pilot Management and Application.
It is necessary links of autonomous driving vehicle operation that autonomous driving vehicle independently refuels, and autonomous driving vehicle
A key technology.How to allow autonomous driving vehicle independently to drive to realize at oil gun during oiling is automatic Pilot sport technique segment
The problem for needing to solve.Traditional automatic Pilot path design mostly is artificially defined driving path, e.g., for this gas station
Environment, sets corresponding track.But this method motility and adaptability be not strong.In reality, due to the design of gas station
Structure is different, and environment is different, the oiling travel route being manually set often less effective.
The content of the invention
A kind of autonomous oiling method of autonomous driving vehicle and intelligent vehicle is embodiments provided, to realize intelligent vehicle
Autonomous traveling is refueled.The technical scheme is as follows：
On the one hand, there is provided a kind of autonomous oiling method of autonomous driving vehicle, methods described include：
The traveling strategy value table of the intelligent vehicle, the traveling plan is formed by intelligent vehicle autonomic learning using learning algorithm
Slightly value table is formed by each the grid assignment in the multiple grids segmented to running region；
Based on the traveling strategy value table, the travel route between the intelligent vehicle and the refueling position is planned automatically；
According to the travel route automatic running to the refueling position, to be refueled.
Autonomous driving vehicle provided in an embodiment of the present invention independently travels oiling method, calculates using classical machine learning
Method-Q learning algorithms, by autonomous driving vehicle autonomic learning, realize under varying environment that (driving path is most for optimum traveling strategy
It is few) so that autonomous driving vehicle while realizing that autonomous traveling is refueled, with more preferable stability, adaptivity, mobility
And motility.
Alternatively, in one embodiment, the utilization learning algorithm forms the intelligence by intelligent vehicle autonomic learning
The traveling strategy value table of car includes：
The state of the intelligent vehicle is initialized, and a decision-making action is randomly generated for the state；
State to reaching after the decision-making action judges；
If not up to described refueling position, the maximum step number for not up to arranging do not collide, repeat to randomly generate
The step of state that decision-making action fighting to the finish is reached after instigating to make is judged；
If reaching the refueling position, reaching the maximum step number of setting or collide, the row of the intelligent vehicle is updated
Sail strategy value table, and carry out next time study test until reaching default test number (TN).
Alternatively, in another embodiment, methods described also includes：
When default test number (TN) is reached, last time test is updated the traveling strategy value table for obtaining and normalizes to 0
And between 1, as final traveling strategy value table.
Alternatively, in one embodiment, it is described based on the traveling strategy value table, plan automatically the intelligent vehicle and institute
The travel route stated between refueling position includes：
Direction according to value from small to large on the traveling strategy value table forms the intelligent vehicle with the oiling position
Travel route between putting.
Alternatively, in various embodiments of the present invention, the travel route between the intelligent vehicle and the refueling position
It is planning, to be provided with the intelligent vehicle special short according to the distance between the fuel tank cap of the intelligent vehicle and gas station's oil gun
Journey communication technology DSRC module.
On the other hand, there is provided a kind of intelligent vehicle, the intelligent vehicle include：Study module, route planning module and traveling mould
Block, wherein：
The study module, for forming the traveling plan of the intelligent vehicle using learning algorithm by intelligent vehicle autonomic learning
Slightly it is worth table, the traveling strategy value table is formed by each the grid assignment in the multiple grids segmented to running region；
The layout of roads module, for based on the traveling strategy value table, planning that automatically the intelligent vehicle is added with described
Oil level put between travel route；
The traveling module, for according to the travel route automatic running to the refueling position, to be refueled.
Alternatively, in one embodiment, the study module specifically for：
The state of the intelligent vehicle is initialized, and a decision-making action is randomly generated for the state；
State to reaching after the decision-making action judges；
If not up to described refueling position, the maximum step number for not up to arranging do not collide, repeat to randomly generate
The step of state that decision-making action fighting to the finish is reached after instigating to make is judged；
If reaching the refueling position, reaching the maximum step number of setting or collide, the row of the intelligent vehicle is updated
Sail strategy value table, and carry out next time study test until reaching default test number (TN).
Alternatively, in another embodiment, the study module is additionally operable to：
When default test number (TN) is reached, by last time test update traveling strategy value table normalize to 0 and 1 it
Between, as final traveling strategy value table.
Alternatively, in another embodiment, the layout of roads module specifically for：
Direction according to value from small to large on the traveling strategy value table forms the intelligent vehicle with the oiling position
Travel route between putting.
Alternatively, in various embodiments of the present invention, DSRC technology DSRC is installed on the intelligent vehicle
Module, the layout of roads module specifically for：According to the distance between the fuel tank cap of the intelligent vehicle and gas station's oil gun come
Plan the travel route between the intelligent vehicle and the refueling position.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought is：
The traveling strategy value table of intelligent vehicle is formed by autonomic learning algorithm, and then carries out automatically route planning, can be made
Route running of the intelligent vehicle based on planning is simultaneously independently refueled, and realizes the intellectuality refueled.
Description of the drawings
For the technical scheme being illustrated more clearly that in the embodiment of the present invention, below will be to making needed for embodiment description
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, can be obtaining other according to these accompanying drawings
Accompanying drawing.
Fig. 1 is a kind of flow chart of the autonomous oiling method of autonomous driving vehicle provided in an embodiment of the present invention；
Fig. 2 is the schematic diagram of the learning process of machine learning algorithm provided in an embodiment of the present invention；
Fig. 3 is the flow chart based on Q learning algorithms provided in an embodiment of the present invention；
Fig. 4 a-4d are the schematic diagrams of Q learning algorithms value table more new principle provided in an embodiment of the present invention；
Fig. 5 is the schematic diagram that all parts in the embodiment of the present invention in autonomous driving vehicle are arranged；
Fig. 6 is to realize the autonomous schematic diagram for refueling based on Q learning algorithms in the embodiment of the present invention；
Fig. 7 is the structured flowchart of intelligent vehicle provided in an embodiment of the present invention.
Specific embodiment
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is described in further detail.
Fig. 1 is a kind of flow chart of the autonomous oiling method of autonomous driving vehicle provided in an embodiment of the present invention.With reference to Fig. 1,
The autonomous oiling method of autonomous driving vehicle provided in an embodiment of the present invention may include：
11st, the traveling strategy value table of the intelligent vehicle, the row is formed using learning algorithm by intelligent vehicle autonomic learning
Sail strategy value table to be formed by each the grid assignment in the multiple grids for segmenting running region.
Wherein, the learning algorithm can be classical machine learning algorithm-Q learning algorithms.Value table is the table with value
Running region is subdivided into multiple grids by lattice, by each grid imparting value, you can obtain value table.
Wherein, form the traveling plan of the intelligent vehicle described in this step using learning algorithm by intelligent vehicle autonomic learning
Slightly value table includes：The state of the intelligent vehicle is initialized, and a decision-making action is randomly generated for the state；It is dynamic to the decision-making
The state reached after work is judged；If not up to described refueling position, the maximum step number for not up to arranging do not collide,
Then repeat to randomly generate decision-making action and fight to the finish instigate to make after the state that reaches the step of judged；If reaching the oiling position
Put, reach the maximum step number of setting or collide, then update the traveling strategy value table of the intelligent vehicle, and learned next time
Test is practised until reaching default test number (TN).
Wherein, the decision-making action in the embodiment of the present invention may include：Steering wheel left-hand rotation angle, right-hand rotation angle, throttle dynamics,
Braking strength etc..
Alternatively, in one embodiment, the autonomous oiling method of autonomous driving vehicle provided in an embodiment of the present invention may be used also
Including：When default test number (TN) is reached, last time test is updated the traveling strategy value table for obtaining and normalizes to 0 and 1
Between, as final traveling strategy value table.
12nd, based on the traveling strategy value table, plan automatically the traveling road between the intelligent vehicle and the refueling position
Line.
Alternatively, it is described based on the traveling strategy value table, planned between the intelligent vehicle and the refueling position automatically
Travel route may include：Direction according to value from small to large on the traveling strategy value table forms the intelligent vehicle and institute
State the travel route between refueling position.
In embodiments of the present invention, the travel route between the intelligent vehicle and the refueling position is according to the intelligence
The distance between the fuel tank cap of car and gas station's oil gun are planning.
Meanwhile, DSRC technology (Dedicated Short Range can be installed on the intelligent vehicle
Communications, DSRC) module.
13rd, according to the travel route automatic running to the refueling position, to be refueled.
The embodiment of the present invention forms the traveling strategy value table of intelligent vehicle by autonomic learning algorithm, and then carries out automatically route
Planning, can make route running of the intelligent vehicle based on planning and independently be refueled, realize the intellectuality refueled.
Fig. 2 is the schematic diagram of the learning process of machine learning algorithm provided in an embodiment of the present invention.With reference to Fig. 2, drive automatically
Sail automobile by with the environment real-time, interactive for not having priori, perceive current state amount X (t) of system, and make one
Decision-making action u (t).This decision-making action can change current ambient conditions so that system reaches a new quantity of state X (t+1).
At the same time, environment can feed back to one enhancing signal r (t) of autonomous driving vehicle, to represent that autonomous driving vehicle decision-making is moved
Make the return immediately of u (t).Generally, enhancing signal is numerically present, and different numerical value are to evaluate and distinguish decision-making action
" good ", " bad ".Equally, for new quantity of state X (t+1), autonomous driving vehicle can make new decision-making action u (t+1) again,
And new enhancing signal r (t+1) is fed back to from environment.Go down by that analogy, i.e., autonomous driving vehicle can at each moment
With environmental interaction, by " good " of the enhancing signal of environmental feedback, " bad ", on-line control decision strategy is carried out, so as to follow-up
The return of maximum is obtained in decision-making action.
On the basis of theory of algorithm, the disclosure can be learnt first on computers, and the process of study is, first to automatic
Driving initiation parameter is designed, and random initializtion autonomous driving vehicle state simultaneously randomly generates one certainly to the state
Instigate to make, the state that reaches after instigating to make of fighting to the finish carries out logical judgment, if still miss the mark, less than maximum step number, repeat
Randomly generate decision-making action；No, then updated value table, is updated to value table by lot of experiments, as autonomous driving vehicle most
The experience of excellent driving path.These experiences are loaded into into real vehicle, are implemented to transfer with environmental interaction and experience by vehicle, is realized most
Excellent autonomous oiling path policy.
In embodiments of the present invention, the auto-oiled travel route of autonomous driving vehicle based on Q learning algorithms was planned
Journey is a learning process, and autonomous driving vehicle learns how solid with most short driving path arrival in success with the experience of failure
Determine refueling position.Because there is failure phenomenon in learning process, therefore needing first to be tested on computers (learn), treating automatic Pilot
After automobile succeeds in school, algorithm can be transplanted on real vehicle (that is, intelligent vehicle).
In embodiments of the present invention, autonomous driving vehicle is independently driven to up to the refueling position specified, and is really driven automatically
The fuel tank cap position for sailing automobile reaches and apart from the position of oil gun respective distance.Therefore, fuel tank cap can be regarded as a particle (automatically
Driving position), oil gun position regards a particle (target location) as.Autonomous refueling can be converted into local path
Planning problem.Path planning problem is solved using Q learning algorithms in the embodiment of the present invention.
Fig. 3 is the flow chart based on Q learning algorithms provided in an embodiment of the present invention.With reference to Fig. 3, the embodiment of the present invention is provided
The process of Q learning algorithms may include：Algorithm parameter initialization design, update original state sequence, randomly generate decision-making action,
Collision judgment, dbjective state judge, step number judges and are worth table updates.Specifically can be as follows：
Algorithm parameter initialization design
It is, for example, 100 to arrange maximum test number (TN) (MaxTrail), and maximum mobile step number (MaxStep) is, for example, 7, initially
Test number (TN) (Trail) is 0, reach target and obtain enhancing signal r=+1, collide, obtain enhancing signal r=-1, other
State obtains enhancing signal r=0.
Update original state sequence
Each time on-test, the mobile step number (step) of initialization is 0, randomly selects original state (i.e. position).
Randomly generate decision-making action
At original state (position), a decision-making action will be randomly generated, reach a new state (adjacent states).
Clash logic judges
If colliding, updated value table, and next test is carried out, the process of original state sequence is updated again；
If not colliding, proceed to next step.
Dbjective state judges
If reaching dbjective state, updated value table, and next test is carried out, be updated the mistake of original state sequence again
Journey；If not reaching dbjective state, proceed to next step.
Step number judges
Judge vehicle movement step number whether more than maximum mobile step number.If vehicle movement step number is more than maximum mobile step number,
Then updated value table, and next test is carried out, the process of original state sequence is updated again；If vehicle movement step number is not more than
Maximum mobile step number, then carry out randomly generating the process of decision-making action again.
It is pointed out that in the embodiment of the present invention, clash logic judges, dbjective state judges and step number judges to press
Perform according to the order for arranging above, i.e. these three judgements can be without clear and definite sequencing, can select as needed in practice
Execution sequence.For example, it is also possible to perform according to dbjective state judgement, the order that clash logic judges and step number judges etc..
Value table updates
Fig. 4 a-4d are the schematic diagrams of Q learning algorithms value table more new principle provided in an embodiment of the present invention.Herein with figure
A simply example shown in 4a-4d updates rule with new annotating Q learning algorithm value tables.In Fig. 4 a-4d, 9 grids represent intelligence
Possible state is planted in the 9 of energy robot place environment presence, and arrow is the action that intelligent robot may be selected, and is represented from one
Individual state is shifted to the state of another state.The increasing that digital expression state in figure beside each arrow-action transfer is obtained
Strong signal.Q-value represents from one action of a condition selecting largest cumulative enhancing signal for obtaining, and V-value represents that a state can
With the largest cumulative enhancing signal for obtaining.G represents dbjective state, once will stop in this condition into this state.Fig. 4
In (a), under original state, except the Q-value that can reach target is that, in addition to 100, the Q-value of other state transfers is 0.In Fig. 4 (b),
Commutation factor α=0.8 is chosen, according to formulaIntelligent robot is carried out instead to each state
Again after execution action, update Q-value.Wherein, x represents current state, and x ' represents NextState, and u represents current action, u ' tables
Show next action, r is enhancing signal,Represent optimum working value.In Fig. 4 (c), according to formula
Obtain the corresponding V functions of each state, wherein, V*X () represents optimum state value.Fig. 4 (d) is represented according to Q functions and V functions
The optimal route (i.e. no matter from the beginning of which state, reach in target step number minimum) that obtains of numerical value.And so on, until all
Off-test (that is, when current step number is more than maximum step number).The value table that last time test updates can be normalized to [0,1],
As optimal strategy value table.
Algorithm (optimal strategy value table) after study is loaded into real vehicle, as the Jing of autonomous driving vehicle optimal route selection
After testing, automobile is capable of achieving automatic circuit planning, completes autonomous oiling.
Fig. 5 is the schematic diagram that all parts in the embodiment of the present invention in autonomous driving vehicle are arranged.Autonomous driving vehicle
Sensor is installed as shown in Figure 5.Autonomous driving vehicle front, rear, left and right can be separately installed with a millimetre-wave radar 51 (in figure
One is illustrated by way of example only), for detecting surrounding enviroment, a DSRC is installed (specially at autonomous driving vehicle fuel tank cap
With short-range communication technology) module 52, the oil gun of gas station's oil tank is provided with DSRC modules 52, realizes logical between vehicle and oil gun
Letter, and then know the position (target) of oil gun.DSRC is a kind of efficient wireless communication technology, is utilized in the embodiment of the present invention
DSRC can realize the identification and two-way communication in the domain of specific cell to the mobile target under high-speed motion
Fig. 6 is to realize the autonomous schematic diagram for refueling based on Q learning algorithms in the embodiment of the present invention.During actual travel,
Service station environment is carried out into sliding-model control so that autonomous oiling navigational challenge is changed into the path planning problem of Q learning algorithms.Profit
With optimal strategy value table, the optimal path of discrete environment is calculated, recycle autonomous driving vehicle driving path to take fitting discrete
Optimal path, and then autonomous driving vehicle is independently driven at oil gun mouth.With reference to Fig. 6, between intelligent vehicle 61 and oil tank 63
Travel route be planning according to the distance between the reservoir port 62 of the intelligent vehicle 61 and gas station's oil gun 64.Show in Fig. 6
A kind of vehicle line cooked up according to the embodiment of the present invention is gone out.
The autonomous oiling method of autonomous driving vehicle provided in an embodiment of the present invention, is calculated with classical machine learning algorithm-Q study
Method, by autonomous driving vehicle autonomic learning, realizes that optimum traveling is tactful (driving path is minimum) under varying environment so that automatically
Driving is autonomous to refuel with more preferable stability, adaptivity, mobility and motility.
Fig. 7 is the structured flowchart of intelligent vehicle provided in an embodiment of the present invention.With reference to Fig. 7, intelligence provided in an embodiment of the present invention
Energy car 700 may include：Study module 701, route planning module 702 and traveling module 703.Wherein：
The study module 701, for forming the row of the intelligent vehicle using learning algorithm by intelligent vehicle autonomic learning
Strategy value table is sailed, the traveling strategy value table is by each the grid assignment shape in the multiple grids segmented to running region
Into；
The layout of roads module 702, for based on it is described traveling strategy value table, plan automatically the intelligent vehicle with it is described
Travel route between refueling position；
The traveling module 703, for according to the travel route automatic running to the refueling position, to carry out adding
Oil.
Intelligent vehicle provided in an embodiment of the present invention, and then route planning is carried out automatically, intelligent vehicle can be made based on planning
Route running is simultaneously independently refueled, and realizes the intellectuality refueled.
Alternatively, the study module 701 can be specifically for：The state of the intelligent vehicle is initialized, and is directed to the state
Randomly generate a decision-making action；State to reaching after the decision-making action judges；If not up to described refueling position, not
Reach the maximum step number of setting or do not collide, then repeat to randomly generate decision-making action and fight to the finish instigate to make after the state that reaches
The step of being judged；If reaching the refueling position, reaching the maximum step number of setting or collide, the intelligence is updated
The traveling strategy value table of car, and carry out next time study test until reaching default test number (TN).
Alternatively, the study module 701 can be additionally used in：
When default test number (TN) is reached, by last time test update traveling strategy value table normalize to 0 and 1 it
Between, as final traveling strategy value table.
Alternatively, the layout of roads module 702 can be specifically for：
Direction according to value from small to large on the traveling strategy value table forms the intelligent vehicle with the oiling position
Travel route between putting.
In various embodiments of the present invention, DSRC technology DSRC module is installed on the intelligent vehicle 700.
In various embodiments of the present invention, alternatively, the layout of roads module 702 can be specifically for：According to described
The distance between the fuel tank cap of intelligent vehicle and gas station's oil gun are planning the traveling between the intelligent vehicle and the refueling position
Route.
Intelligent vehicle provided in an embodiment of the present invention, using classical machine learning algorithm-Q learning algorithms, by driving automatically
Automobile autonomic learning is sailed, realizes that optimum traveling is tactful (driving path is minimum) under varying environment so that autonomous driving vehicle is autonomous
Refuel with more preferable stability, adaptivity, mobility and motility.
It should be noted that：The intelligent vehicle that above-described embodiment is provided only carries out illustrating with the division of above-mentioned each functional module
It is bright, in practical application, can as desired by above-mentioned functions distribution be completed by different functional modules, will intelligent vehicle it is interior
Portion's structure is divided into different functional modules, to complete all or part of function described above.In addition, above-described embodiment is carried
For intelligent vehicle and the autonomous oiling method embodiment of autonomous driving vehicle belong to same design, which implements process and refers to method
Embodiment, is repeated no more here.
Unless otherwise defined, technical term used herein or scientific terminology should be in art of the present invention and have
The ordinary meaning understood by the personage of general technical ability.Used in present patent application description and claims " the
One ", " second " and similar word are not offered as any order, quantity or importance, and are used only to distinguish different
Ingredient.Equally, the similar word such as " one " or " " does not indicate that quantity is limited yet, but represents and have at least one.
The word that " connection " or " being connected " etc. are similar to is not limited to physics or machinery connection, and can be including electrical
Connection, either directly still indirectly.
One of ordinary skill in the art will appreciate that realizing that all or part of step of above-described embodiment can pass through hardware
To complete, it is also possible to instruct the hardware of correlation to complete by program, described program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read only memory, disk or CD etc..
The foregoing is only presently preferred embodiments of the present invention, not to limit the present invention, all spirit in the present invention and
Within principle, any modification, equivalent substitution and improvements made etc. should be included within the scope of the present invention.
Claims (10)
1. the autonomous oiling method of a kind of autonomous driving vehicle, it is characterised in that methods described includes：
The traveling strategy value table of the intelligent vehicle, the traveling strategy value is formed by intelligent vehicle autonomic learning using learning algorithm
Table is formed by each the grid assignment in the multiple grids segmented to running region；
Based on the traveling strategy value table, the travel route between the intelligent vehicle and the refueling position is planned automatically；
According to the travel route automatic running to the refueling position, to be refueled.
2. method according to claim 1, it is characterised in that the utilization learning algorithm passes through intelligent vehicle autonomic learning shape
Traveling strategy value table into the intelligent vehicle includes：
The state of the intelligent vehicle is initialized, and a decision-making action is randomly generated for the state；
State to reaching after the decision-making action judges；
If not up to described refueling position, the maximum step number for not up to arranging do not collide, repeat to randomly generate decision-making
The step of state that action fighting to the finish is reached after instigating to make is judged；
If reaching the refueling position, reaching the maximum step number of setting or collide, the traveling plan of the intelligent vehicle is updated
Slightly be worth table, and carry out next time study test until reaching default test number (TN).
3. method according to claim 2, it is characterised in that methods described also includes：
When default test number (TN) is reached, by last time test update the traveling strategy value table that obtains normalize to 0 and 1 it
Between, as final traveling strategy value table.
4. method according to claim 1, it is characterised in that described based on the traveling strategy value table, plans institute automatically
The travel route stated between intelligent vehicle and the refueling position includes：
On the traveling strategy value table direction according to value from small to large formed the intelligent vehicle and the refueling position it
Between travel route.
5. according to the arbitrary described method of claim 1-4, it is characterised in that between the intelligent vehicle and the refueling position
Travel route is planning, to install on the intelligent vehicle according to the distance between the fuel tank cap of the intelligent vehicle and gas station's oil gun
There is DSRC technology DSRC module.
6. a kind of intelligent vehicle, it is characterised in that the intelligent vehicle includes：Study module, route planning module and traveling module, its
In：
The study module, for forming the traveling strategy value of the intelligent vehicle using learning algorithm by intelligent vehicle autonomic learning
Table, the traveling strategy value table are formed by each the grid assignment in the multiple grids segmented to running region；
The layout of roads module, for based on the traveling strategy value table, planning automatically the intelligent vehicle with the oiling position
Travel route between putting；
The traveling module, for according to the travel route automatic running to the refueling position, to be refueled.
7. intelligent vehicle according to claim 6, it is characterised in that the study module specifically for：
The state of the intelligent vehicle is initialized, and a decision-making action is randomly generated for the state；
State to reaching after the decision-making action judges；
If not up to described refueling position, the maximum step number for not up to arranging do not collide, repeat to randomly generate decision-making
The step of state that action fighting to the finish is reached after instigating to make is judged；
If reaching the refueling position, reaching the maximum step number of setting or collide, the traveling plan of the intelligent vehicle is updated
Slightly be worth table, and carry out next time study test until reaching default test number (TN).
8. intelligent vehicle according to claim 7, it is characterised in that the study module is additionally operable to：
When default test number (TN) is reached, the traveling strategy value table that last time test updates is normalized between 0 and 1,
As final traveling strategy value table.
9. intelligent vehicle according to claim 6, it is characterised in that the layout of roads module specifically for：
On the traveling strategy value table direction according to value from small to large formed the intelligent vehicle and the refueling position it
Between travel route.
10. according to the arbitrary described intelligent vehicle of claim 6-9, it is characterised in that be provided with special short distance on the intelligent vehicle
Communication technology DSRC module, the layout of roads module specifically for：According to fuel tank cap and gas station's oil gun of the intelligent vehicle
The distance between planning the travel route between the intelligent vehicle and the refueling position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710024566.1A CN106525063A (en) | 2017-01-11 | 2017-01-11 | Autonomous refueling method of autonomous car and intelligent car |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710024566.1A CN106525063A (en) | 2017-01-11 | 2017-01-11 | Autonomous refueling method of autonomous car and intelligent car |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106525063A true CN106525063A (en) | 2017-03-22 |
Family
ID=58337025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710024566.1A Pending CN106525063A (en) | 2017-01-11 | 2017-01-11 | Autonomous refueling method of autonomous car and intelligent car |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106525063A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107123197A (en) * | 2017-04-12 | 2017-09-01 | 深圳市赛亿科技开发有限公司 | The selfoiling system and oiling method of a kind of automatic driving vehicle |
| CN107300388A (en) * | 2017-06-04 | 2017-10-27 | 东南大学 | Tourism route planing method of riding based on Q learning algorithms and echo state network |
| CN107416754A (en) * | 2017-07-21 | 2017-12-01 | 北京图森未来科技有限公司 | A kind of automatic oiling methods, devices and systems of long-distance vehicle |
| CN109076761A (en) * | 2018-08-08 | 2018-12-25 | 安徽省现代农业装备产业技术研究院有限公司 | One kind being based on remote sensing images variable fertilization method |
| CN109496288A (en) * | 2017-07-13 | 2019-03-19 | 北京嘀嘀无限科技发展有限公司 | System and method for determining track |
| CN109884997A (en) * | 2017-07-21 | 2019-06-14 | 北京图森未来科技有限公司 | A kind of vehicle control device and vehicle |
| CN110824912A (en) * | 2018-08-08 | 2020-02-21 | 华为技术有限公司 | Method and apparatus for training a control strategy model for generating an autonomous driving strategy |
| US11756022B2 (en) | 2017-06-16 | 2023-09-12 | Nokia Technologies Oy | Methods, apparatus and computer programs for enabling transactions using digital attributes |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140306814A1 (en) * | 2013-04-15 | 2014-10-16 | Flextronics Ap, Llc | Pedestrian monitoring application |
| CN105083278A (en) * | 2015-07-31 | 2015-11-25 | 奇瑞汽车股份有限公司 | Vehicle control method and device |
| CN105139072A (en) * | 2015-09-09 | 2015-12-09 | 东华大学 | Reinforcement learning algorithm applied to non-tracking intelligent trolley barrier-avoiding system |
| CN105128856A (en) * | 2015-08-24 | 2015-12-09 | 奇瑞汽车股份有限公司 | Method and device for parking vehicle into garage |
| CN105346483A (en) * | 2015-11-04 | 2016-02-24 | 常州加美科技有限公司 | Man-machine interactive system for unmanned vehicle |
| CN105912814A (en) * | 2016-05-05 | 2016-08-31 | 苏州京坤达汽车电子科技有限公司 | Lane change decision model of intelligent drive vehicle |
| CN105928531A (en) * | 2016-04-13 | 2016-09-07 | 浙江合众新能源汽车有限公司 | Method for generating route accurately used for pilotless automobile |
| CN106114507A (en) * | 2016-06-21 | 2016-11-16 | 百度在线网络技术（北京）有限公司 | Local path planning method and device for intelligent vehicle |
-
2017
- 2017-01-11 CN CN201710024566.1A patent/CN106525063A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140306814A1 (en) * | 2013-04-15 | 2014-10-16 | Flextronics Ap, Llc | Pedestrian monitoring application |
| CN105083278A (en) * | 2015-07-31 | 2015-11-25 | 奇瑞汽车股份有限公司 | Vehicle control method and device |
| CN105128856A (en) * | 2015-08-24 | 2015-12-09 | 奇瑞汽车股份有限公司 | Method and device for parking vehicle into garage |
| CN105139072A (en) * | 2015-09-09 | 2015-12-09 | 东华大学 | Reinforcement learning algorithm applied to non-tracking intelligent trolley barrier-avoiding system |
| CN105346483A (en) * | 2015-11-04 | 2016-02-24 | 常州加美科技有限公司 | Man-machine interactive system for unmanned vehicle |
| CN105928531A (en) * | 2016-04-13 | 2016-09-07 | 浙江合众新能源汽车有限公司 | Method for generating route accurately used for pilotless automobile |
| CN105912814A (en) * | 2016-05-05 | 2016-08-31 | 苏州京坤达汽车电子科技有限公司 | Lane change decision model of intelligent drive vehicle |
| CN106114507A (en) * | 2016-06-21 | 2016-11-16 | 百度在线网络技术（北京）有限公司 | Local path planning method and device for intelligent vehicle |
Non-Patent Citations (2)
| Title |
|---|
| 中国人工智能学会: "《中国人工智能学会第10届全国学术年会论文集 下 中国人工智能进展》", 30 November 2003, 北京邮电大学出版社 * |
| 文孟飞等: "《城市智能交通系统控制与诱导的理论与实践》", 30 November 2015, 中南大学出版社 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107123197A (en) * | 2017-04-12 | 2017-09-01 | 深圳市赛亿科技开发有限公司 | The selfoiling system and oiling method of a kind of automatic driving vehicle |
| CN107300388A (en) * | 2017-06-04 | 2017-10-27 | 东南大学 | Tourism route planing method of riding based on Q learning algorithms and echo state network |
| CN107300388B (en) * | 2017-06-04 | 2020-05-08 | 东南大学 | Riding tour route planning method based on Q-learning algorithm and echo state network |
| US11756022B2 (en) | 2017-06-16 | 2023-09-12 | Nokia Technologies Oy | Methods, apparatus and computer programs for enabling transactions using digital attributes |
| CN109496288A (en) * | 2017-07-13 | 2019-03-19 | 北京嘀嘀无限科技发展有限公司 | System and method for determining track |
| US11708261B2 (en) | 2017-07-21 | 2023-07-25 | Beijing Tusen Weilai Technology Co., Ltd. | Method, device and system for automatic oiling of long-distance transport vehicle |
| CN107416754A (en) * | 2017-07-21 | 2017-12-01 | 北京图森未来科技有限公司 | A kind of automatic oiling methods, devices and systems of long-distance vehicle |
| CN109884997A (en) * | 2017-07-21 | 2019-06-14 | 北京图森未来科技有限公司 | A kind of vehicle control device and vehicle |
| US10800645B2 (en) | 2017-07-21 | 2020-10-13 | Tusimple, Inc. | Method, device and system for automatic oiling of long-distance transport vehicle |
| CN109884997B (en) * | 2017-07-21 | 2021-01-08 | 北京图森智途科技有限公司 | Vehicle controller and vehicle |
| US11720103B2 (en) | 2017-07-21 | 2023-08-08 | Beijing Tusen Zhitu Technology Co., Ltd. | System, method and associated apparatus for autonomous vehicle transportation |
| US11467577B2 (en) | 2017-07-21 | 2022-10-11 | Beijing Tusen Zhitu Technology Co., Ltd. | System, method and associated apparatus for autonomous vehicle transportation |
| CN109076761A (en) * | 2018-08-08 | 2018-12-25 | 安徽省现代农业装备产业技术研究院有限公司 | One kind being based on remote sensing images variable fertilization method |
| CN109076761B (en) * | 2018-08-08 | 2021-01-08 | 安徽省现代农业装备产业技术研究院有限公司 | Variable fertilization method based on remote sensing image |
| CN110824912A (en) * | 2018-08-08 | 2020-02-21 | 华为技术有限公司 | Method and apparatus for training a control strategy model for generating an autonomous driving strategy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106525063A (en) | Autonomous refueling method of autonomous car and intelligent car | |
| CN107284441B (en) | The energy-optimised management method of the adaptive plug-in hybrid-power automobile of real-time working condition | |
| CN107264534B (en) | Based on the intelligent driving control system and method for driver experience's model, vehicle | |
| CN105549597B (en) | A kind of unmanned vehicle dynamic path planning method based on environmental uncertainty | |
| CN102292752B (en) | Row running control system and vehicle | |
| CN110304074A (en) | A kind of hybrid type driving method based on stratification state machine | |
| CN107168303A (en) | A kind of automatic Pilot method and device of automobile | |
| CN105083278B (en) | Control method for vehicle and device | |
| CN113291308B (en) | Vehicle self-learning lane-changing decision-making system and method considering driving behavior characteristics | |
| CN106427987A (en) | System and method for controlling hybrid vehicle | |
| CN104167097B (en) | A kind of generation method of the system of path generator of dynamically overtaking other vehicles based on truck traffic | |
| CN110297494A (en) | A kind of automatic driving vehicle lane-change decision-making technique and system based on rolling game | |
| CN107886750B (en) | Unmanned automobile control method and system based on beyond-visual-range cooperative cognition | |
| CN106926844A (en) | A kind of dynamic auto driving lane-change method for planning track based on real time environment information | |
| CN106845716A (en) | A kind of unmanned surface vehicle local delamination paths planning method based on navigation error constraint | |
| CN107577234A (en) | A kind of automobile fuel ecomomy control method of driver in ring | |
| CN110992695A (en) | Intelligent vehicle urban intersection traffic decision multi-objective optimization model based on conflict resolution | |
| CN105109482A (en) | Parking garage entry method and device | |
| CN108694841A (en) | A kind of intelligent vehicle passage crossroads traffic light method based on V2X technologies | |
| CN113291318B (en) | Unmanned vehicle blind area turning planning method based on partially observable Markov model | |
| CN106843210A (en) | One kind is based on bionic automatic driving vehicle progress control method | |
| CN105109480B (en) | Intelligent turning method and device | |
| CN113312760B (en) | Network online motor vehicle right-turn track planning method and device based on traffic simulation | |
| Zöldy et al. | Cognitive mobility–cogmob | |
| CN110320916A (en) | Consider the autonomous driving vehicle method for planning track and system of occupant's impression |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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: 20170322 |
|
| RJ01 | Rejection of invention patent application after publication |