CN110364026A - A kind of vehicle follow-up strategy safe verification method and system based on state reachable set - Google Patents
A kind of vehicle follow-up strategy safe verification method and system based on state reachable set Download PDFInfo
- Publication number
- CN110364026A CN110364026A CN201910735536.0A CN201910735536A CN110364026A CN 110364026 A CN110364026 A CN 110364026A CN 201910735536 A CN201910735536 A CN 201910735536A CN 110364026 A CN110364026 A CN 110364026A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- state
- follow
- strategy
- reachable set
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000012795 verification Methods 0.000 title claims abstract description 35
- 238000013459 approach Methods 0.000 claims abstract description 15
- 238000012216 screening Methods 0.000 claims abstract description 7
- 231100001261 hazardous Toxicity 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000012546 transfer Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 5
- 230000007704 transition Effects 0.000 claims description 5
- 238000012790 confirmation Methods 0.000 claims description 4
- 238000012512 characterization method Methods 0.000 claims description 3
- 230000003750 conditioning effect Effects 0.000 claims description 3
- 230000001394 metastastic effect Effects 0.000 claims description 3
- 206010061289 metastatic neoplasm Diseases 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 238000012502 risk assessment Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 description 12
- 238000011217 control strategy Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000001133 acceleration Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003542 behavioural effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 241000736199 Paeonia Species 0.000 description 1
- 235000006484 Paeonia officinalis Nutrition 0.000 description 1
- NPYPAHLBTDXSSS-UHFFFAOYSA-N Potassium ion Chemical compound [K+] NPYPAHLBTDXSSS-UHFFFAOYSA-N 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 229910001414 potassium ion Inorganic materials 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- G06Q50/40—
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
Abstract
The invention belongs to vehicle follow-up strategy technical fields, disclose a kind of vehicle follow-up strategy safe verification method and system based on state reachable set, vehicle behavior is described using the state traversals performance characteristic of reachable set, and hazardous act is provided by reachable set intersection, the screening for carrying out early warning and security control scheme, finally obtains safe follow-up strategy;Again in such a way that Markov Chain approaches reachable set expression system do not know behavior state variation, further verify the validity of safe follow-up strategy.Present invention expression system in such a way that Markov Chain approaches reachable set does not know the variation of behavior state, verifies the validity of safe follow-up strategy;And modeling implementation method proposed by the present invention not only imperfectly characterizes and examines vehicle and do not know behavior, and complete safe model- following control decision scheme can also be provided, significantly improve the applicability of follow-up strategy.
Description
Technical field
The invention belongs to vehicle follow-up strategy technical fields more particularly to a kind of vehicle based on state reachable set to follow
Security policy verification method and system.
Background technique
Currently, the immediate prior art:
The driver of vehicle is followed to guarantee its driving safety by a variety of strategy of speed control, these strategies follow one
Fixed safety criterion such as keeps certain safe spacing with front truck, or speed control is such as the following in preceding vehicle speed.It is many
There is the building for following model to embody these basic concepts, however these follow modeling to be only capable of exporting a kind of safe model- following control,
And the safety knot opinion of certain single determining control strategy can only be also exported based on the verifying of motion model.It is built it can be seen that these are followed
Mould method substantially belongs to the prediction of the vehicle behavior under specific model- following control decision, can neither reflect all feasible safety controls
System strategy, can not also cover all possible system action dynamic change of vehicle, thus can not theoretically guarantee to vehicle with
With the completeness and accuracy of policy validation.
For this purpose, the present invention is calculated by the accessibility being distributed to vehicle-state, realize dynamic to vehicle behavior under follow-up strategy
The complete expression and parsing of state.Be different from the statement of conventional truck action trail, the motion state distribution of vehicle be it is various not
It determines the statement under disturbance to vehicle behavior state trajectory DYNAMIC DISTRIBUTION, is from vehicle original state set to object of planning position
Set a kind of probabilistic assurance of all state trajectories of set.This statement is since vehicle is transported with the basic foundation held
There is uncertainties for input, the output of dynamic system, and vehicle, which is one, in addition has Random Discrete control and real-time continuous behavior
The complication system of mode mixture superimposed characteristics, resulting in the actual motion state of vehicle, there is complicated uncertainties.However
Previous modeling method is completely stated this uncertainty and is difficult, because the behavior dynamic of single track portrays deficiency
To show this uncertainty comprehensively.And Formal Modeling is exactly based on the reachable set modeling of traversal vehicle behavior state
Mode can not only clearly describe the dynamic system structure of vehicle and correlation properties, and be capable of the complete of automatic Ergodic Theory
State space, thus the completeness and credibility that theoretically state uncertainty characterizes under support vehicles follow-up strategy.
Therefore, the security verification of vehicle follow-up strategy can be converted into the accessibility of state of motion of vehicle under the strategy
Analysis, key are the reachable set modeling and effectively calculating of state of motion of vehicle.Mainly pass through currently, reachable set calculates
Excessively approximate (Over-Approximation) in system mode domain is abstract, such as convex polyhedron, Piecewise affine systems, ellipsoid
Approximate morphologic appearance originates from the form of system current state field, iterates to calculate the variation of subsequent form to realize.Due to mistake
The abstract computation complexity of approximation is high, cost is big, can not analyze high-dimensional complication system, therefore mixed to general nonlinearity
It closes the performance on automatic machine and is not so good as people's will.Therefore under the reachable set modeling conditions expressed based on convex polyhedron, mostly with more
The numerical value form of expression of the face body as linear hybrid automaton basic status domain, most representative is Althoff et al. to vehicle
State reachable set analysis method.
Safe follow-up strategy is first it is ensured that vehicle will not collide with front truck in model- following control, and secondly speed can be with
Control retains certain space headway within front truck rate limitation, or with front vehicles.In addition, meeting above-mentioned condition
On the basis of, the final limiting safe condition of additional authentication vehicle control is still needed to, i.e., in emergency case inferoanterior vehicle and Following Car
Emergency brake is until safety stop simultaneously.
In conclusion problem of the existing technology is:
(1) conventional truck follow modeling analysis be able to validate only vehicle single behavior safety, and cannot be limited
Secondary operation can traverse the uncertainty of vehicle all inputs and operating status.Complete peace theoretically cannot be disposably provided
Full model- following control decision.
(2) driver is not one kind accurately with stable behavior pattern to the practical control of vehicle, existing to follow model
Obtaining control result is a kind of stringent accurate control list entries, and operability is poor.
(3) the existing behavior safety verification method calculated based on model can only provide qualitative conclusions, such as dangerous or safety,
Fail safety or risk to driving behavior and makes quantitative statistical analysis.
Solve the difficulty of above-mentioned technical problem:
It cannot achieve the sampling simulation calculation to all planning control schemes of driver in short time.
Sampling emulation not can guarantee the result covering to vehicle control state edge data.
Driving behavior habit can not follow model that must plan to agree with tradition.
It can not carry out optimal, suboptimum of control strategy etc. without being perfectly safe in the traffic situation in channel and divide Hierarchical Decision Making.
Solve the meaning of above-mentioned technical problem:
It can realize to control driver in the calculating of limited times based on the Formal Modeling that strict mathematical defines and go
For comprehensive analysis verifying;
The breakthrough disposable confirmation to decision is followed in the past is realized and follows the strategy expression of decision set to safety and divide
Analysis;
In conjunction with existing calculation method, it can be achieved that the online quick application of strategy, has high actual utility.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of vehicle follow-up strategies based on state reachable set
Safe verification method and system.The present invention is current by being judged by vehicle-state reachable set intersection based on up to set analysis
The safety of control strategy follows traveling to provide plurality of optional decision thus to obtain security control set for vehicle safety.This
Outside, it is the uncertainty of further expression vehicle following behavior, vehicle following state reachable set approximation is abstracted as Markov
Chain by expressing the uncertain changing rule of vehicle following state to markovian update, and verifies safety of the invention
The validity of follow-up strategy modeling method.
The invention is realized in this way a kind of vehicle follow-up strategy safe verification method based on state reachable set, described
Vehicle follow-up strategy safe verification method based on state reachable set includes:
The first step first with the status and appearance signature analysis vehicle behavior of reachable set, and is provided by reachable set intersection
Vehicle risk behavior, carry out early warning and security control scheme screening, finally obtain safe follow-up strategy.
Then in such a way that Markov Chain approaches reachable set expression system do not know behavior state variation, into one
Step demonstrate,proves the validity of safe follow-up strategy.
Further, it in the first step, is analyzed in vehicle behavior using the state traversals performance characteristic of reachable set, including vehicle
State reachable set characterization.
Behavior dynamic model under vehicle nondeterministic statement is established using the method that reachable set approximation characterizes, gives one kind
Vehicle kinematics modelWhereinIndicate the motion state of vehicle, u ∈ U indicates the control of vehicle
Input.As t=r, accurate vehicle-state reachable set Re(t) it calculates as follows:
Wherein, X (0) is initial state space.Using road longitudinal spaceWith speedBuilding
System continuous state space X=[S, V], the corresponding continuous control input space areX (0) ∈ X in above formula
(0)=[S (0), V (0)], at this time under original state vehicle reachable set Re(r) by the two-dimensional state of lengthwise position and speed sky
BetweenApproximate representation, i.e.,
R is acquired using super approximate calculation methode(r) approximation R (r), so thatCorrespondingly, certain a period of time
Between in range the reachable set of vehicle be R [0, r]=∪t∈[0,r]R(t)。
Further, the first step, and vehicle risk behavior is provided by reachable set intersection, carry out early warning and security control side
In the screening of case, the restrictive condition in follow-up strategy is set as corresponding safe judgement event, is specifically included:
Event 1 follows vehicle and the space headway between vehicle is guided to reach early warning value, is presented as t moment, two parking stalls
The relationship set between state reachable set isIts
Middle F and L is respectively that vehicle and front is followed to guide vehicles identifications, and D is space headway early warning value,To sum it up operation,Expression will follow the position reachable set of vehicle to extend D distance forward.
Event 2 follows vehicle to stop traveling, is presented as current time t, follows car speed vF(t)=0.
Event 3 follows speed to be equal to guidance speed, is presented as t moment, vF(t)=vL(t), v (t)=center (Rv
(t)), center () is cluster center, Rv(t) velocity information in state reachable set is indicated.If vL(t)-vF(t) value is by bearing
Increase is zero, and trigger event 3 is positive triggering, is otherwise negative sense triggering.
Event 4 follows vehicle to collide with guidance vehicle, is presented as that current time t, two truck position reachable sets intersect,
I.e.
Further, in the first step, the method for obtaining safe follow-up strategy includes:
1) system inputs: { SF(0), SL(0), VF(0), VL(0), UF(0), UL(0), T }.
2) t=0: whether event 1 triggers: being that primary condition is precarious position, exports risk judgment P=0.It is no, just
Beginning condition is safe condition, into next checking procedure.
3) [0, T] t=: vehicle and guidance vehicle reachable set is followed to calculate.
4) t=t*∈ [0, T]: trigger event 1.If uF(0)>uL(0) (u (0)=center (U (0))), P=0.
If uF(0)<uL(0), t*=T, vF(t*)>vL(t*), P=0.
vF(t*)<vL(t*), P=1.
t*< T, t '=[t*, T], trigger event 1 or event 2 or event 3.
P=0.Trigger event 4 triggers, P=0 without event.
5) t=t*∈ [0, T]: trigger event 2, P=1.
6) t=t*∈ [0, T]: trigger event 3.
If t*=T, P=1;
If t*< T, t '=[t*, T], trigger event 2 or without event trigger, P=1.Trigger event 1 or 4, P=0.
7) t=t*∈ [0, T]: trigger event 4, P=0.
8) [0, T] t ∈: no event triggering enables
uL(0)=- 1, T '=100*T.
T=[0, T ']: vehicle and guidance vehicle reachable set is followed to calculate.
T ∈ [0, T ']: trigger event 4, P=0.Otherwise, P=1.
Wherein, P=0 indicates that current control is dangerous, and P=1 indicates current control safety.
Further, the method for obtaining safe follow-up strategy further comprises:
After system continuous state space and input space discretization, Vehicular system discrete space X is obtainedi,j=[Si,Vj],
I=1,2 ..., n, j=1,2 ..., m and Uα, α=1,2 ..., g.Vehicle is followed in correlating event and guides the first of vehicle
Beginning state and control input are respectively A=(zF=(i, j), yF=α), B=(zL=(i', j'), yL=β), then follow vehicle
Safety traffic probability are as follows:
Wherein, P (T) indicates the probability of happening that control input duration is T, determines Following Car in safe follow-up strategy
Current time selects the probability of a certain discrete control inflow section αAre as follows:
Further, in second step, expression system does not know behavior state in such a way that Markov Chain approaches reachable set
Variation, in the validity for further verifying safe follow-up strategy, specifically include:
Step 1, discretization vehicle-state reachable set, and approached with predictable Markov Chain.Estimated by following formula
Vehicle is calculated in the Markov Chain of subsequent time and period:
Wherein, Γ (tk) it is that feasible time-varying control inputs transfer matrix, reflect control input in safe follow-up strategy
Select statistical probability.Moment state-transition matrix Φ (τ) and period state-transition matrix Φ (0, τ), wherein τ is Ma Erke
The time step of husband's chain.
Step 2, work as tkWhen moment vehicle-state is X=(i, j), follow vehicle control input by β section to α section
Transition probability isThe then probability of vehicle selection control input α section are as follows:
Wherein,
Wherein, Ψ is intrinsic transition matrix, is the control input variation rule in the intrinsic behavior of vehicle.Parameter γ value is got over
Greatly, vehicle control input adjustment is more frequent, and the smaller then vehicle control of value inputs fewer adjustment.λ is priority variable, reflection
When car speed is restricted by traffic environment to passively take speed change measure, vehicle is excellent to control input discrete segment
First select probability.In an intrinsic section, extreme driving condition is then selected in intrinsic section the preferential selection of vehicle control input
Outside, the probability that control inputs preferential selection section α is denoted as μα, it is in the vehicle control input selection for the state of being traveled freely generally
Rate is
Step 3, vehicle preferentially selects to control the probability of inflow section α under safe follow-up strategyWith intrinsic probability μαFor
Basis, conditional probabilityFor the upper limit, is converted and is determined according to such as following formula:
Wherein,In formula (8), in order to meet condition:Separately have
Further, second step in the validity for verifying safe follow-up strategy, passes through building off-line calculation and online verification
Computational frame models the random reachable set of state of motion of vehicle, shows the uncertainty with vehicle individual behavior generation of speeding.
In the off-line calculation, first with the continuous movement of the Discrete Change approximate expression vehicle of state reachable set, to vehicle
The manifestation mode that the initial random process reachable set of motion state uses Markov Chain to approach, will initial continuous vehicle
If motion state space is divided into dry lattice, discretization vehicle movement shape space is carried out;Then computing system state is by a column grid
Lattice are transferred to the probability of another column grid, and result is stored into Markov transferring matrix.
Then, the metastatic rule of approximate abstract state of motion of vehicle variation, extracts state of motion of vehicle and state shifts square
Battle array.
Again by controlling the safe with state relation matrix between vehicle under the mode of speeding of input offer in line computation, drawing
Under guide-car conditioning for provided with vehicle safety decision of speeding modular operation selection gist and confirmation select probability;And
And by verifying to the off-line simulation with behavior safety between associated vehicle in the mode of speeding, vehicle-state and control input are obtained
Between security association matrix.
Finally, the trace of vehicle is guided to be distributed the update iteration for realizing state space according to gained Markov chain;It follows
Vehicle is constraint with incidence matrix according to guidance vehicle and itself current status condition, and the risk for obtaining different operation selection is commented
Valence carries out the preferred of safety approach.
Another object of the present invention is to provide the vehicle follow-up strategy safety described in a kind of implementation based on state reachable set
The vehicle follow-up strategy security authentication systems based on state reachable set of verification method.
Another object of the present invention is to provide the vehicle follow-up strategy safety described in a kind of realize based on state reachable set
The information data processing terminal of verification method.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer
When upper operation, so that computer executes the vehicle follow-up strategy safe verification method based on state reachable set.
In conclusion advantages of the present invention and good effect are as follows:
Markov Chain proposed by the present invention based on vehicle-state reachable set approaches to realize safe follow-up strategy modeling
With the method for verifying.This method describes vehicle behavior using the state traversals performance characteristic of reachable set, and passes through reachable set intersection
The early warning of hazardous act and the screening of security control scheme are provided, to disposably completely embody safe follow-up strategy;Then
Expression system does not know the variation of behavior state in such a way that Markov Chain approaches reachable set, verifies safe follow-up strategy
Validity.The experimental results showed that the modeling implementation method of proposition, which not only imperfectly characterizes and examines vehicle, does not know row
For complete safe model- following control decision scheme can also be provided, significantly improve the safety and applicability of follow-up strategy.
Detailed description of the invention
Fig. 1 is the vehicle follow-up strategy safe verification method process provided in an embodiment of the present invention based on state reachable set
Figure.
Fig. 2 is that t ∈ [2-2.5] s provided in an embodiment of the present invention follows the feasible input of vehicle, speed and position histogram.
Fig. 3 is the vehicle trace distribution map of different time section provided in an embodiment of the present invention.
Fig. 4 is the collision probability figure under safe follow-up strategy provided in an embodiment of the present invention between vehicle.
Fig. 5 is Technology Roadmap of the present invention when realizing application on site.
Fig. 6 is program runtime data of the embodiment of the present invention under application on site method and technology route (in embodiment
Method is applied at line computation time (0.007s)) figure.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to this hair
It is bright to be further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, not
For limiting the present invention.
Conventional truck follow modeling analysis be able to validate only vehicle single behavior safety, and cannot be in limited times
Operation can traverse the uncertainty of vehicle all inputs and operating status.Complete safety theoretically cannot be disposably provided
Model- following control decision.
In view of the problems of the existing technology, the present invention provides a kind of vehicle follow-up strategies based on state reachable set
Safe verification method is with reference to the accompanying drawing explained in detail the present invention.
As shown in Figure 1, the vehicle follow-up strategy safety verification side provided in an embodiment of the present invention based on state reachable set
Method, comprising:
S101 describes vehicle behavior using the state traversals performance characteristic of reachable set, and provided by reachable set intersection
Hazardous act carries out the screening of early warning and security control scheme, finally obtains safe follow-up strategy.
S102, expression system does not know the variation of behavior state in such a way that Markov Chain approaches reachable set, into one
Step demonstrate,proves the validity of safe follow-up strategy.
The invention will be further described combined with specific embodiments below.
Embodiment
Vehicle follow-up strategy safe verification method provided in an embodiment of the present invention based on state reachable set includes:
1 system modelling.
1.1 modeling method.
In complicated road traffic environment, follows the motion state of vehicle to change and there is complicated uncertainty, these
Uncertainty comes not only from the random fluctuation of behavior state in vehicle operation, be also derived from by front vehicles position with
Control selections is uncertain caused by speed influences, therefore following vehicle following the trace in decision to be distributed (is various
It is from vehicle original state set to the object of planning to the statement of vehicle behavior state trajectory DYNAMIC DISTRIBUTION under uncertain disturbance
A kind of probabilistic assurance of all state trajectories of location sets) change with very strong random nature.For this purpose, proposing to adopt
With the idea about modeling of reachable set, the safety of the vehicle model- following control strategy of expressed intact and parsing under various condition of uncertainty
Property.
Essentially, the safety issue of vehicle follow-up strategy is a kind of multilevel security decision problem, passes through building thus
The safety of multilayer event decision model parsing model- following control.In this model, it is assumed that follow between vehicle and guidance vehicle
Behavior is mutually indepedent, and respectively to two vehicle traces in the model- following control period point under current motion state and control input condition
The reachable set of cloth models, and calculates the probability of fore-aft vehicle trace distribution reachable set relativeness, analyzes model- following control strategy
Degree of risk, verifying follow vehicle current all feasible control strategies safety and safe coefficient.Effectively to count
The application effect of safe follow-up strategy follows all possible safety control strategy of vehicle to select according to Principle of Statistics simulation,
It is inputted according to certain all feasible security controls of statistical probability selection, rather than the control of a certain determination inputs.
1.2 system modelling processes.
1) vehicle-state reachable set characterizes.
For the uncertain behavior state for expressing vehicle, vehicle is established using the method that reachable set approximation characterizes and does not know shape
Behavior dynamic model under state, detailed modeling process is referring to early-stage study.In brief, a kind of vehicle kinematics model is givenWhereinIndicate the motion state of vehicle, u ∈ U indicates the control input of vehicle.As t=r,
Accurate vehicle-state reachable set Re(t) it calculates as follows:
Wherein, X (0) is initial state space.The present invention uses road longitudinal spaceWith speedBuilding system continuous state space X=[S, V], the corresponding continuous control input space areTherefore
X (0) ∈ X (0)=[S (0), V (0)] in above formula, at this time under original state vehicle reachable set Re(r) can by lengthwise position with
The two-dimensional state space of speedApproximate representation, i.e.,
The present invention uses super approximate calculation method[11-12]Acquire Re(r) approximation R (r), so thatPhase
Ying Di, sometime the reachable set of vehicle can be approximately R [0, r]=∪ in ranget∈[0,r]R(t)。
2) follow-up strategy multistage event is stated.
Restrictive condition in follow-up strategy is set as corresponding safe judgement event, is specifically included:
Event 5 follows vehicle and the space headway between vehicle is guided to reach early warning value, is presented as t moment, two truck positions
Relationship between state reachable set isWherein F
It is respectively that vehicle and front is followed to guide vehicles identifications with L, D is space headway early warning value,To sum it up operation,Expression will follow the position reachable set of vehicle to extend D distance forward.
Event 6 follows vehicle to stop traveling, is presented as current time t, follows car speed vF(t)=0.
Event 7 follows speed to be equal to guidance speed, is presented as t moment, vF(t)=vL(t), v (t)=center (Rv
(t)), " center () " is cluster center, Rv(t) velocity information in state reachable set is indicated.If vL(t)-vF(t) value by
Negative increase is zero, trigger event 3, referred to as positive triggering, is otherwise negative sense triggering, it is clear that forward direction triggering is security control, is born
It need to be further confirmed that triggering, the present invention only considers that negative sense triggers.
Event 8 follows vehicle to collide with guidance vehicle, is presented as that current time t, two truck position reachable sets intersect,
I.e.
The model- following control safety verification process that logical order according to the triggering of each event is determined is as shown in table 1:
1 model- following control safety verification process of table
Note: P=0 indicates that current control is dangerous, and P=1 indicates current control safety.
Vehicle system is obtained after system continuous state space and input space discretization for the uncertainty of expression system
Unite discrete space Xi,j=[Si,Vj], i=1,2 ..., n, j=1,2 ..., m and Uα, α=1,2 ..., g.If definition association
Vehicle is followed in event and guides the original state and control input respectively A=(z of vehicleF=(i, j), yF=α), B=(zL
=(i', j'), yL=β), then follow the safety traffic probability of vehicle are as follows:
Wherein, P (T) indicates that the probability of happening that control input duration is T, value are obtained without going through detection, can be according to
Its numerical value is set according to the situation that is evenly distributed of driver.Thus it can determine and vehicle current time followed to select in safe follow-up strategy
Select the probability of a certain discrete control inflow section αAre as follows:
3) safe follow-up strategy verifying.
The uncertainty of safe follow-up strategy under being examined for access control, discretization vehicle-state reachable set, and with can
The Markov Chain of prediction is approached.Here markovian update depends on two transfer matrixes: the transfer of moment state
Matrix Φ (τ) and period state-transition matrix Φ (0, τ), wherein τ is markovian time step.For this purpose, under passing through
Formula estimates vehicle in the Markov Chain of subsequent time and period:
Wherein, Γ (tk) it is that feasible time-varying control inputs transfer matrix, reflect control input in safe follow-up strategy
Select statistical probability.Work as tkWhen moment vehicle-state is X=(i, j), if following vehicle control input by β section to α section
Transition probability beThe then probability of vehicle selection control input α section are as follows:
Wherein,
Wherein, Ψ is intrinsic transition matrix, i.e., the control input variation rule in the intrinsic behavior of vehicle shows as vehicle control
Random jump of the system input between discrete segment, jump difference between control input (or control inflow section sequential digit values it
Difference) it is bigger, jump may be smaller;Parameter γ value is bigger, and vehicle control input adjusts more frequent, the smaller then vehicle of value
Control inputs fewer adjustment;λ is priority variable, and reflection car speed is restricted passively to take change by traffic environment
When fast measure, preferential select probability of the vehicle to control input discrete segment.For the universal driving habit of the mankind, vehicle control
The preferential selection of system input (habit section) in an intrinsic section, only extreme driving condition then selects outside intrinsic section,
The probability of the preferential selection section α of this control input is denoted as μα, that is, it is in the vehicle control input selection for the state of being traveled freely generally
Rate is
Finally, vehicle preferentially selects to control the probability of inflow section α under safe follow-up strategyWith intrinsic probability μαFor
Basis, conditional probabilityFor the upper limit, determined according to such as down conversion:
Wherein,In formula (8), in order to meet condition:Separately have
Below with reference to simulating, verifying, the invention will be further described.
Vehicle behavior is limited on specified reference orbit by emulation, i.e., assuming that transverse path deviates and longitudinal movement
Under the premise of mutually indepedent, to longitudinal direction of car (direction along ng a path) and lateral behavior Independent modeling.The simplified kinematics of longitudinal movement
Model is described, and motion profile is indicated by trace distribution probability f (s), and cross track, which deviates, then uses segmentation possibilities distribution function
F (δ) approximate expression.Then, the trace distribution probability in vehicle future is represented by f (s, δ)=f (s) f (δ).In addition, in view of
Trace is distributed visual purpose, constructs the distribution of vehicle body referring to vehicle entity width.
The expression formula of vehicle longitudinal movement is as follows:
The major parameter of the motion model has vehicle defeated along the position s of track, speed v and a standardization acceleration
Enter control parameter u, the constant interval of the parameter is [- 1,1], and -1 indicates that vehicle is braked with all strength, and 1 indicates to accelerate with all strength, by tire
The absolute extremes acceleration a that frictional force is limitedmaxWith constant v*It is determined by the characteristic property of different traffic participants, this
Invention puts aside negotiation of bends environment.In order to more clearly from show the vehicle trace potassium ion distribution in lane, by same vehicle
More vehicles are separately displayed in different lanes with trace distribution at any time in road, to prevent different vehicle in visualization result
Trace is overlapped the bring collimation error.
For the validity for verifying vehicle safety follow-up strategy, it is assumed that A in Through Lane, B are arranged successively after tri- Chinese herbaceous peony of C same
To traveling, the parking stall A in direction of travel forefront, the initial velocities of three vehicles according to the successive incremented by successively of direction of travel, i.e. A < B <
Constant C, A vehicle speed is 3m/s, and three vehicle vehicles are identical, are wide 2m, long 5m.During vehicle behavior characteristic and verifying calculate
Major parameter assignment is shown in Table 2, and initial attribute is shown in Table 3.
2 behavioral trait of table and major parameter
3 initial attribute of table: it is uniformly distributed set
Travel situations of three vehicles in 8s in the traffic situation are simulated.2-2.5s Following Car in time section
Fig. 2 is shown in feasible acceleration control input selection and corresponding speed and change in location under safety control strategy;Three vehicle tracks
The visualization result of mark is shown in Fig. 3 by Time segments division, and the distribution color of each vehicle is normalized respectively in figure, deep in figure
Color region indicates high probability distribution, and light areas indicates low probability distribution;Collision probability between three vehicles is shown in Fig. 4.
As seen from Figure 2, safe follow-up strategy can select to follow vehicle B, C to provide various control decision scheme, vehicle
Corresponding safe speed and in-position can also be distributed in different discrete state sections, and by track between three vehicles in Fig. 3
The calculating that mark intersects probability just obtains the validity statistics of the safe follow-up strategy of the present invention, as shown in figure 4, since vehicle A moves shape
State is stablized, and vehicle B is smaller with its collision probability in the process of moving, and vehicle C-state then changes with the fluctuation of vehicle B, therefore two
Risk of collision between vehicle is higher than the former.It can be seen that the extreme influence that the uncertainty of vehicle behavior follows safety, also prints
The validity and necessity of the safe follow-up strategy examined the present invention is based on reachable set calculating are demonstrate,proved.
Vehicle A and the collision probability of vehicle B are minimum in Fig. 4, and vehicle B and the collision probability of vehicle C are larger, to find out its cause, the former draws
It leads vehicle behavior to stablize, the latter guides vehicle for the former to follow vehicle, and behavior uncertainty is earlier above, it is seen that when following vehicle
When can not know the accurate behavior of guidance vehicle, tradition follows decision-making technique to be unable to satisfy and determines to the analysis of uncertain behavior
Plan, this is also exactly the design aim of follow-up strategy modeling implementation method of the present invention, this result embodies the method for the present invention to true
It is qualitative to follow and the uncertain general applicability followed.
Fig. 5 is Technology Roadmap of the present invention when realizing application on site.
In order to improve vehicle safety with the efficiency for mode online verification of speeding, off-line calculation and online verification calculation block are constructed
The model of frame, system action is constituted and logical relation is as shown, dotted line is linked as numerical value update and its direction in figure.
The off-line calculation first purpose of model framework is provided for online verification with speeding needed for state of motion of vehicle estimation
Multidate information (offline part one).Then, it is modeled by the random reachable set to state of motion of vehicle, performance is with vehicle of speeding
The uncertainty that body behavior is generated by various disturbances.The present invention is with the Discrete Change approximate expression vehicle of state reachable set
Continuous movement, for this purpose, the performance side for using Markov Chain to approach the initial random process reachable set of state of motion of vehicle
Formula, and this approximate procedure is carried out in two steps: if initial continuous system state space (1) is divided into dry lattice, with this from
Dispersion system state space;(2) computing system state is transferred to the probability of another column grid by a column grid, and result is stored up
It is stored in Markov transferring matrix.With the metastatic rule of the abstract state of motion of vehicle variation of this approximation, vehicle movement shape is extracted
State and state-transition matrix, to reduce on-line operation difficulty, reduce On-line Estimation time it is worth mentioning that the calculating only
It is correlation behavior influence abstract with the approximation for state of motion of vehicle changing rule of speeding, not being related between vehicle.
The second purpose is to provide safety for On-line Control input with state relation matrix between vehicle under the mode of speeding, so as to
In the case where guiding vehicle conditioning for provided with vehicle safety decision of speeding modular operation selection gist and confirmation select probability
(offline part two).For this purpose, obtaining vehicle by verifying to the off-line simulation with behavior safety between associated vehicle in the mode of speeding
Security association matrix between state and control input.
In on-line analysis part, guides the trace of vehicle to be distributed and realize state space more according to gained Markov chain
New iteration;It follows vehicle according to guidance vehicle and itself current status condition, is constraint with incidence matrix, different operation can be obtained
The risk assessment of selection, so that it is preferred to carry out safety approach.
Fig. 6 is program runtime data of the embodiment of the present invention under application on site method and technology route (in embodiment
Method is applied at line computation time (0.007s)) figure.
Below with reference to specific effect, the invention will be further described.
Tradition follows modeling and analysis methods that can only export the single trace information for following vehicle or its safety knot opinion, thus
Both behavioral trait can have been known to vehicle and has carried out complete characterization, also fail to reflect all feasible security control decisions.For this purpose,
It is proposed the vehicle safety follow-up strategy modeling and analysis methods based on state reachable set.
Experiment shows: the method for proposition can not only traverse the holonomic system state space that vehicle follows decision automatically,
The limitation to uncertain behavior expression is overcome, expressible vehicle follows driving status, sensor measurement and auto model
Deng uncertain behavior state caused by uncertainty, model is improved to the expressive force of vehicle agenda, is theoretically ensured
The completeness of vehicle follow-up strategy security verification.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or
Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to
Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network
Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one
A computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from
One web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line
(DSL) or wireless (such as infrared, wireless, microwave etc.) mode is into another web-site, computer, server or data
The heart is transmitted).The computer-readable storage medium can be any usable medium that computer can access either
The data storage devices such as server, the data center integrated comprising one or more usable mediums.The usable medium can be
Magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
Solid State Disk (SSD)) etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of vehicle follow-up strategy safe verification method based on state reachable set, which is characterized in that it is described based on state can
Vehicle follow-up strategy safe verification method up to collection includes:
The first step analyzes vehicle behavior, and the vehicle provided by reachable set intersection using the state traversals performance characteristic of reachable set
Hazardous act carries out the screening of early warning and security control scheme, finally obtains safe follow-up strategy;
Second step, expression system does not know the variation of behavior state in such a way that Markov Chain approaches reachable set, further
Verify the validity of safe follow-up strategy.
2. the vehicle follow-up strategy safe verification method based on state reachable set as described in claim 1, which is characterized in that the
It in one step, is analyzed in vehicle behavior using the state traversals performance characteristic of reachable set, including vehicle-state reachable set characterization;
Behavior dynamic model under vehicle nondeterministic statement is established using the method that reachable set approximation characterizes, gives a kind of vehicle fortune
It is dynamic to learn modelWhereinIndicate the motion state of vehicle, u ∈ U indicates the control input of vehicle;Work as t
When=r, accurate vehicle-state reachable set Re(t) it calculates as follows:
Wherein, X (0) is initial state space;Using road longitudinal spaceWith speedBuilding system
Continuous state space X=[S, V], the corresponding continuous control input space areX (0) ∈ X (0)=[S in above formula
(0), (0) V], at this time under original state vehicle reachable set Re(r) by the two-dimensional state space of lengthwise position and speedApproximate representation, i.e.,
R is acquired using super approximate calculation methode(r) approximation R (r), so thatCorrespondingly, sometime range
The reachable set of interior vehicle is R [0, r]=∪t∈[0,r]R(t)。
3. the vehicle follow-up strategy safe verification method based on state reachable set as described in claim 1, which is characterized in that the
One step, the vehicle risk behavior provided by reachable set intersection carry out that plan will be followed in early warning and the screening of security control scheme
Restrictive condition in slightly is set as corresponding safe judgement event, specifically includes:
Event 1 follows vehicle and the space headway between vehicle is guided to reach early warning value, is presented as t moment, two truck position states
Relationship between reachable set isWherein F and L points
Vehicle and front Wei not be followed to guide vehicles identifications, D is space headway early warning value,To sum it up operation,Indicating will
The position reachable set of vehicle is followed to extend D distance forward;
Event 2 follows vehicle to stop traveling, is presented as current time t, follows car speed vF(t)=0;
Event 3 follows speed to be equal to guidance speed, is presented as t moment, vF(t)=vL(t), v (t)=center (Rv(t)),
Center () is cluster center, Rv(t) velocity information in state reachable set is indicated;If vL(t)-vF(t) value is by negative increase
Zero, trigger event 3 is positive triggering, is otherwise negative sense triggering;
Event 4 follows vehicle to collide with guidance vehicle, is presented as that current time t, two truck position reachable sets intersect, i.e.,
4. the vehicle follow-up strategy safe verification method based on state reachable set as described in claim 1, which is characterized in that the
In one step, the method for obtaining safe follow-up strategy includes:
1) system inputs: { SF(0), SL(0), VF(0), VL(0), UF(0), UL(0), T };
2) t=0: whether event 1 triggers: being that primary condition is precarious position, exports risk judgment P=0;It is no, primary condition
For safe condition, into next checking procedure;
3) [0, T] t=: vehicle and guidance vehicle reachable set is followed to calculate;
4) t=t*∈ [0, T]: trigger event 1;If uF(0)>uL(0) (u (0)=center (U (0))), P=0;
If uF(0)<uL(0), t*=T, vF(t*)>vL(t*), P=0;
vF(t*)<vL(t*), P=1;
t*< T, t '=[t*, T], trigger event 1 or event 2 or event 3;
P=0;Trigger event 4 triggers, P=0 without event;
5) t=t*∈ [0, T]: trigger event 2, P=1;
6) t=t*∈ [0, T]: trigger event 3;
If t*=T, P=1;
If t*< T, t '=[t*, T], trigger event 2 or without event trigger, P=1;Trigger event 1 or 4, P=0;
7) t=t*∈ [0, T]: trigger event 4, P=1;
8) [0, T] t ∈: no event triggering enables
uF(0)=- 1;
uL(0)=- 1, T '=100*T;
T=[0, T ']: vehicle and guidance vehicle reachable set is followed to calculate;
T ∈ [0, T ']: trigger event 4, P=0;Otherwise, P=1;
Wherein, P=0 indicates that current control is dangerous, and P=1 indicates current control safety.
5. the vehicle follow-up strategy safe verification method based on state reachable set as claimed in claim 4, which is characterized in that obtain
The method for taking safe follow-up strategy further comprises:
After system continuous state space and input space discretization, Vehicular system discrete space X is obtainedi,j=[Si,Vj], i=
1,2 ..., n, j=1,2 ..., m and Uα, α=1,2 ..., g;Vehicle is followed in correlating event and guides the initial shape of vehicle
State and control input are respectively A=(zF=(i, j), yF=α), B=(zL=(i', j'), yL=β), then follow the safety of vehicle
Travel probability are as follows:
Wherein, P (T) indicates the probability of happening that control input duration is T, determines in safe follow-up strategy and follows vehicle current
Moment selects the probability of a certain discrete control inflow section αAre as follows:
6. the vehicle follow-up strategy safe verification method based on state reachable set as described in claim 1, which is characterized in that the
In two steps, expression system does not know the variation of behavior state in such a way that Markov Chain approaches reachable set, further verifies
In the validity of safe follow-up strategy, comprising the following steps:
Step 1, discretization vehicle-state reachable set, and approached with predictable Markov Chain;It is estimated by following formula
Markov Chain of the vehicle in subsequent time and period:
Wherein, Γ (tk) it is that feasible time-varying control inputs transfer matrix, reflect the selection system that input is controlled in safe follow-up strategy
Count probability;Moment state-transition matrix Φ (τ) and period state-transition matrix Φ (0, τ), when wherein τ is markovian
Between step-length;
Step 2, work as tkWhen moment vehicle-state is X=(i, j), follow vehicle control input general by the transfer of β section to α section
Rate isThe then probability of vehicle selection control input α section are as follows:
Wherein,
Wherein, Ψ is intrinsic transition matrix, is the control input variation rule in the intrinsic behavior of vehicle;Parameter γ value is bigger, vehicle
Control input adjustment is more frequent, and the smaller then vehicle control of value inputs fewer adjustment;λ is priority variable, reflection vehicle speed
When degree is restricted by traffic environment to passively take speed change measure, vehicle is general to the preferential selection of control input discrete segment
Rate;In an intrinsic section, extreme driving condition then selects outside intrinsic section the preferential selection of vehicle control input, will control
The probability of the preferential selection section α of input is denoted as μα, be in the state of being traveled freely vehicle control input select probability be
Step 3, vehicle preferentially selects to control the probability of inflow section α under safe follow-up strategyWith intrinsic probability μαBased on,
Conditional probabilityFor the upper limit, is converted and is determined according to such as following formula:
Wherein,
To meet condition:Separately have
7. the vehicle follow-up strategy safe verification method based on state reachable set as described in claim 1, which is characterized in that the
Two steps, in the validity for verifying safe follow-up strategy, by building off-line calculation and online verification Computational frame, to vehicle movement
The random reachable set of state models, and shows the uncertainty with vehicle individual behavior generation of speeding;
In the off-line calculation, first with the continuous movement of the Discrete Change approximate expression vehicle of state reachable set, vehicle is transported
The manifestation mode that the initial random process reachable set of dynamic state uses Markov Chain to approach, will initial continuous vehicle movement shape
If state space is divided into dry lattice, discretization vehicle movement shape space is carried out;Then computing system state is shifted by a column grid
To the probability of another column grid, and result is stored into Markov transferring matrix;
Then, the metastatic rule of approximate abstract state of motion of vehicle variation, extracts state of motion of vehicle and state-transition matrix;
Again by controlling the safe with state relation matrix between vehicle under the mode of speeding of input offer in line computation, in guidance vehicle
Under conditioning for provided with vehicle safety decision of speeding modular operation selection gist and confirmation select probability;And by pair
With the off-line simulation verifying of behavior safety between associated vehicle in the mode of speeding, the safety between vehicle-state and control input is obtained
Incidence matrix;
Finally, the trace of vehicle is guided to be distributed the update iteration for realizing state space according to gained Markov chain;Follow vehicle
According to guidance vehicle and itself current status condition, it is constraint with incidence matrix, obtains the risk assessment of different operation selection, into
Row safety approach it is preferred.
8. it is a kind of implement claim 1 described in the vehicle follow-up strategy safe verification method based on state reachable set based on state
The vehicle follow-up strategy security authentication systems of reachable set.
9. a kind of vehicle follow-up strategy safety verification side realized described in claim 1~7 any one based on state reachable set
The information data processing terminal of method.
10. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer executes such as
It is required that the vehicle follow-up strategy safe verification method described in 1-7 any one based on state reachable set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910735536.0A CN110364026A (en) | 2019-08-09 | 2019-08-09 | A kind of vehicle follow-up strategy safe verification method and system based on state reachable set |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910735536.0A CN110364026A (en) | 2019-08-09 | 2019-08-09 | A kind of vehicle follow-up strategy safe verification method and system based on state reachable set |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110364026A true CN110364026A (en) | 2019-10-22 |
Family
ID=68223587
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910735536.0A Withdrawn CN110364026A (en) | 2019-08-09 | 2019-08-09 | A kind of vehicle follow-up strategy safe verification method and system based on state reachable set |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110364026A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111007858A (en) * | 2019-12-23 | 2020-04-14 | 北京三快在线科技有限公司 | Training method of vehicle driving decision model, and driving decision determining method and device |
CN111065089A (en) * | 2019-11-05 | 2020-04-24 | 东华大学 | Internet of vehicles bidirectional authentication method and device based on crowd sensing |
CN114613130A (en) * | 2022-02-18 | 2022-06-10 | 北京理工大学 | Driving credibility analysis method in traffic and delivery system |
CN114690635A (en) * | 2022-03-22 | 2022-07-01 | 苏州大学 | State estimation and control method and system of mass spring damping system |
-
2019
- 2019-08-09 CN CN201910735536.0A patent/CN110364026A/en not_active Withdrawn
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111065089A (en) * | 2019-11-05 | 2020-04-24 | 东华大学 | Internet of vehicles bidirectional authentication method and device based on crowd sensing |
CN111007858A (en) * | 2019-12-23 | 2020-04-14 | 北京三快在线科技有限公司 | Training method of vehicle driving decision model, and driving decision determining method and device |
CN114613130A (en) * | 2022-02-18 | 2022-06-10 | 北京理工大学 | Driving credibility analysis method in traffic and delivery system |
CN114613130B (en) * | 2022-02-18 | 2023-05-12 | 北京理工大学 | Driving credibility analysis method in traffic and carrying system |
CN114690635A (en) * | 2022-03-22 | 2022-07-01 | 苏州大学 | State estimation and control method and system of mass spring damping system |
CN114690635B (en) * | 2022-03-22 | 2022-10-18 | 苏州大学 | State estimation and control method and system of mass spring damping system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110364026A (en) | A kind of vehicle follow-up strategy safe verification method and system based on state reachable set | |
Bai et al. | Hybrid reinforcement learning-based eco-driving strategy for connected and automated vehicles at signalized intersections | |
Wang et al. | Risk assessment and mitigation in local path planning for autonomous vehicles with LSTM based predictive model | |
KR102461831B1 (en) | System and Method for Improving of Advanced Deep Reinforcement Learning Based Traffic in Non signalalized Intersections for the Multiple Self driving Vehicles | |
US20200278686A1 (en) | Iterative Feedback Motion Planning | |
Ding et al. | Epsilon: An efficient planning system for automated vehicles in highly interactive environments | |
Ward et al. | Probabilistic model for interaction aware planning in merge scenarios | |
Jayawardana et al. | Learning eco-driving strategies at signalized intersections | |
Yang et al. | A multi-state social force based framework for vehicle-pedestrian interaction in uncontrolled pedestrian crossing scenarios | |
Schmidt et al. | An interaction-aware lane change behavior planner for automated vehicles on highways based on polygon clipping | |
Chen et al. | Efficient speed planning for autonomous driving in dynamic environment with interaction point model | |
Ren et al. | Self-learned intelligence for integrated decision and control of automated vehicles at signalized intersections | |
Suriyarachchi et al. | Gameopt: Optimal real-time multi-agent planning and control for dynamic intersections | |
Wu et al. | Motion planning of autonomous vehicles under dynamic traffic environment in intersections using probabilistic rapidly exploring random tree | |
Hu et al. | A roadside decision-making methodology based on deep reinforcement learning to simultaneously improve the safety and efficiency of merging zone | |
Leung et al. | Learning autonomous vehicle safety concepts from demonstrations | |
Ma et al. | Anisotropy safety potential field model under intelligent and connected vehicle environment and its application in car-following modeling | |
Zhu et al. | A decentralized multi-criteria optimization algorithm for multi-unmanned ground vehicles (mugvs) navigation at signal-free intersection | |
Wang et al. | Cooperative decision making for connected automated vehicles in multiple driving scenarios | |
Li et al. | Distributed MPC for multi-vehicle cooperative control considering the surrounding vehicle personality | |
CN107776613A (en) | The flow-optimized control system of subway transportation | |
Cai et al. | General driving behavior model based on the desired safety margin for vehicle flow simulation | |
Xue et al. | Fail-safe behavior and motion planning incorporating shared control for potential driver intervention | |
Takeda et al. | Light Guidance Control of Human Drivers: Driver Modeling, Control System Design, and VR Experiment | |
Naumann | Probabilistic motion planning for automated vehicles |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191022 |