CN110853335A - Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway - Google Patents

Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway Download PDF

Info

Publication number
CN110853335A
CN110853335A CN201911115220.8A CN201911115220A CN110853335A CN 110853335 A CN110853335 A CN 110853335A CN 201911115220 A CN201911115220 A CN 201911115220A CN 110853335 A CN110853335 A CN 110853335A
Authority
CN
China
Prior art keywords
vehicle
decision
vehicles
sequence
cost
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911115220.8A
Other languages
Chinese (zh)
Other versions
CN110853335B (en
Inventor
王立超
杨敏
李斌
徐铖铖
李大韦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN201911115220.8A priority Critical patent/CN110853335B/en
Publication of CN110853335A publication Critical patent/CN110853335A/en
Application granted granted Critical
Publication of CN110853335B publication Critical patent/CN110853335B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles

Abstract

The invention discloses an autonomous decision-making method for collaborative fleet conflict hedging on a common bottleneck section of a highway, which comprises the following specific steps: the method comprises the steps of considering factors such as vehicle time demand intensity, vehicle types and vehicle driving intentions, determining a multi-vehicle conflict risk-avoiding arrangement sequence based on methods that parallel numbers occupy the same sequencing space and are random in probability, establishing a multi-vehicle conflict risk-avoiding behavior decision model based on a MAS system and a countermeasure negotiation mechanism on the basis of the multi-vehicle conflict risk-avoiding arrangement sequence, quantifying an interaction process of each vehicle adjustment intention, determining a decision scheme which meets all vehicle benefit requirements and vehicle driving intention selection preference as much as possible, and realizing autonomous decision of each vehicle conflict risk-avoiding behavior. The invention has the beneficial effects that: the decision-making suggestion is actually provided for the multi-vehicle danger avoiding and releasing process of the common bottleneck highway section, and the technical service support is provided for the further development of the self-adaptive cruise of the automatic driving vehicle.

Description

Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway
Technical Field
The invention relates to an autonomous decision-making method for cooperative fleet conflict avoidance of a frequently-occurring bottleneck road section of an expressway, which is used for autonomous conflict avoidance driving of the cooperative fleet of the frequently-occurring bottleneck road section of the expressway and belongs to the technical field of self-adaptive cruise of automatically-driven vehicles.
Background
The self-adaptive cruise system is an important technical composition of an automatic driving vehicle, and a plurality of self-adaptive cruise driver auxiliary systems are proposed at home and abroad, so that the vehicle can realize autonomous driving and autonomous keeping of a safe distance under a highway with a good driving environment. The current adaptive cruise technology can realize ideal adaptive cruise control under the condition that a single vehicle runs in a good environment, but the current adaptive cruise technology is difficult to realize mutual negotiation and autonomous avoidance of vehicles under the condition of a common bottleneck road section of an expressway, and the collision state among multiple vehicles can seriously influence the adaptive cruise control of the vehicles to cause running danger. The current advanced wireless communication technology and the internet technology provide effective guarantee for interconnection and intercommunication among automatic driving vehicles, and on the basis, the time demand strength, the vehicle types and the vehicle driving intentions between the vehicles in the common bottleneck road section of the expressway are considered, and the interactive co-quotient in the multi-vehicle risk avoidance decision is realized, so that the one-sided and the defect of autonomous driving under the good driving environment of a single vehicle in the existing self-adaptive cruise technology are overcome.
Disclosure of Invention
Aiming at one side and the defects in the prior art, the invention provides an autonomous decision-making method for collision avoidance of a cooperative fleet of frequent bottleneck road sections of a highway, which mainly comprises the following steps:
step 1: and for a plurality of vehicles which have conflicts or potential conflicts, performing conflict resolution sequence initial arrangement by using a method that parallel numbers occupy the same sequencing space according to the time demand intensity of each vehicle.
Step 2: judging whether the initial arrangement of the conflict resolution sequence has the same time requirement strength, if the time requirement strengths are different, determining the initial sequence as a final sequence, and turning to the step 9; if the time-of-existence demand intensity of each vehicle is the same, the sequence goes to step 3.
And step 3: extracting vehicles corresponding to the same time requirement intensity in the initial sequence, and transferring the same time requirement intensity sequence to the step 4; the remaining vehicle time demand intensity sequence remains unchanged and the original time demand intensity initial sequence is transferred to step 8.
And 4, step 4: and (5) carrying out vehicle type priority sequencing on the extracted vehicles with the same time demand intensity, judging whether the vehicles have the same vehicle type under the same time demand intensity, if so, turning to the step 5, and if not, turning to the step 7.
And 5: and extracting the same vehicle type sequence under the same time demand intensity, and transferring the sequence to the step 6.
Step 6: and (5) performing intention interaction on vehicles of the same vehicle type under the same time requirement, renumbering the initial sequence of each vehicle according to the driving intention of the vehicle, and transferring the updating result to the step 8.
And 7: and updating the initial sequence of the demand intensity of the vehicles at the same time according to different types of the vehicles, and transferring the updating result to the step 8.
And 8: and (5) performing sequence updating again on the initial sequence of the remaining vehicle time demand intensity in the step (3) and the updated sequences transferred in the steps (6) and (7) to obtain a final sequence of conflict resolution, and transferring the sequence to a step (9).
And step 9: and (5) on the basis of the final sequence of conflict resolution, carrying out multi-vehicle group negotiation decision by combining the vehicle intention, determining each vehicle behavior adjusting mode, and turning to the step 10.
Step 10: and adjusting the running behavior of each vehicle.
In the step 1, the conflict risk avoiding sequence is initially arranged by using a method of occupying the same sequencing space by parallel numbers according to the time demand intensity of each vehicle, wherein the time demand intensity of the ith vehicle for automatic driving is represented as tau, and tau is provided
Figure BDA0002273851060000021
(i-1, 2,3, …, M.M represents the number of vehicles in the current conflict state, l represents the length of the automatically driven vehicle i from the potential conflict point,v represents the current running speed of the autonomous vehicle i). And comparing the time demand intensity priority levels of the automatic driving vehicles with potential conflicts according to tau, and numbering the automatic driving vehicles in an ascending order from 1 to 1 in a way that the parallel numbers occupy the same sequencing space from high to low levels.
The step 2 of judging whether the same time requirement strength exists in the initial arrangement of the conflict resolution sequence is performed on the basis of the time requirement strength grade number in the step 1, that is, whether the automatic driving vehicles with the same number exist is judged based on the initial sequence of the time requirement strength priority grade.
And 4, judging whether the same vehicle type exists in the extracted vehicles with the same time demand intensity, extracting the time demand intensity sequence in the time demand intensity initial sequence, and comparing whether the vehicle type is the same under the condition of the same time demand intensity.
Step 4-1: and extracting the same time demand intensity sequence, and temporarily marking each vehicle in the sequence to determine each vehicle type.
Step 4-2: and checking whether the types of the vehicles in the same time demand intensity sequence are the same or not, if so, switching to a step 6, and otherwise, switching to a step 7.
In the step 4-1, vehicles corresponding to the same time demand intensity are temporarily marked, and the types of the vehicles are mainly divided into automatic driving vehicles, manual driving vehicles and the like.
In the step 5, the same vehicle type sequence under the same time requirement intensity is extracted, that is, the vehicles with the same vehicle type under the condition of the same time requirement intensity are determined and determined.
In the step 6, vehicles of the same vehicle type under the same time requirement strength are subjected to intention interaction, and the initial sequence of the vehicles is renumbered according to the driving intention of the vehicles, and the specific process is as follows.
Step 6-1: and extracting the same vehicle type sequence under the same time demand intensity and temporarily marking the same vehicle type sequence.
Step 6-2: and generating random unequal adjustment probabilities of the intention of each vehicle.
Step 6-3: and sequencing the random unequal probabilities of the vehicles from small to large.
Step 6-4: and correspondingly adjusting the vehicle arrangement position according to the random unequal probabilities.
Step 6-5: and updating the current sequence according to the random unequal adjustment probability, wherein the updated value of the vehicle sequence corresponding to the minimum value of the random probability is 0, the updated value of the vehicle sequence corresponding to the minimum value of the random probability is 1, and the sequence numbers are all updated by analogy.
The step 6-1 is performed after the comparison and sorting of the time demand intensity and the vehicle type of each vehicle are completed, and here, the vehicles with the same time demand intensity and the same vehicle type are extracted, and the corresponding marks are made on the vehicles corresponding to the extraction sequence.
And 6-2, quantifying the vehicle driving intention by using a random probability method, realizing the vehicle intention interaction process, only considering the adjustment intention intensity of the vehicle in the driving process, generating random unequal probabilities corresponding to each vehicle one by one under the conditions by using a random probability method, and determining the corresponding relation between each random unequal probability and each vehicle.
And 6-3, based on the determination of each vehicle and the corresponding random unequal probability, sequentially arranging the random unequal probabilities from small to large (the larger the probability is, the stronger the contribution intention of the vehicle is).
And 6-4, according to the arrangement sequence of the random unequal probabilities, the corresponding position of the vehicle corresponding to each random unequal probability is changed.
And 6-5, updating the sequence according to the random unequal probability, wherein the vehicle number corresponding to the minimum random probability is 0, the vehicle number corresponding to the next minimum random probability is locked to be 1, and the like is carried out until all the numbers are finished.
And 7: and updating the initial sequence of the demand intensity of the vehicles at the same time according to different types of the vehicles, and transferring the updating result to the step 8.
And 7, sequentially updating the vehicles with the same time demand intensity according to the vehicle type priority, determining a vehicle type priority sequence according to the fact that the reaction time of the automatic driving vehicle in the existing research is lower than that of the manual driving vehicle, and then sequentially numbering the vehicles with the same priority space from high to low in the vehicle type priority sequence and correspondingly increasing the priority level from 0 to 1 according to the vehicle type priority sequence by utilizing the method that the parallel numbering occupies the same sequencing space.
And in the step 8, based on the initial sequence of the time demand intensity of the vehicle, adding the sequence update numbers in the steps 6 and 7 to the initial sequence number to obtain a final collision risk avoiding sequence.
In the step 9, based on the final sequence of conflict resolution, a multi-vehicle group negotiation decision is made in combination with vehicle intentions to determine each vehicle behavior adjustment mode, and the specific process is as follows:
step 9-1: and determining the number corresponding to each vehicle according to the final sequence of the conflict resolution of the plurality of vehicles.
Step 9-2: a selectable adjustment scheme for each vehicle is determined.
Step 9-3: and constructing cost functions to be paid when each vehicle selects the same decision scheme.
Step 9-4: and (4) specifying a payment cost source equation of each vehicle for selecting a scheme.
Step 9-5: a combination of decisions that may be made by each vehicle is determined.
And 9-6: and determining the required payment cost of each vehicle in each decision group.
Step 9-7: and establishing a vehicle driving intention interactive negotiation weight matrix.
And 9-8: and establishing a vehicle driving intention negotiation equation set according to the weight matrix, and solving a negotiation vector.
Step 9-9: and calculating the target total cost corresponding to each group of decision schemes to obtain the minimum total cost, and determining the scheme corresponding to the minimum total cost.
Step 9-10: the cost minimum correspondence scheme is transferred to step 10.
The method for determining the corresponding number of each vehicle according to the final sequence of the multi-vehicle collision risk avoidance in the step 9-1 is m ═ Sort |, wherein m represents the vehicle number, Sort represents the multi-vehicle collision risk avoidance arrangement sequence, and | | represents that the corresponding vehicle number is generated according to the multi-vehicle collision risk avoidance arrangement order.
By using EmAnd (3) a vehicle for carrying out collision avoidance decision of the M (M is 1,2,3, …, M) th vehicle.
Said step 9-2 determining vehicle selectable adjustment using znThe present invention is a method for determining the N (N-1, 2, …, N) th decision-making scheme adopted by the current vehicle, the common adjustment schemes are speed adjustment and lane change adjustment, and the N-1 speed adjustment and the N-2 lane change adjustment are selected as the decision-making scheme of the present invention.
Step 9-3 is to construct a cost function to be paid when each vehicle selects the same decision scheme, wherein the cost function is g (z)n)＝G(g2(zn),g3(zn),…,gM(zn) Wherein z isnRepresents the n-th decision-making scheme, g (z), taken by the vehiclen) Representing the cost function, g, of each vehicle to the nth decision schemem(zn) Represents EmInitial cost for the nth decision scheme, G (G)2(zn),g3(zn),…,gM(zn) Is) represents the comparison of the initial cost of the nth decision scheme for each member in the system.
In the step 9-4, according to the position relationship between the two vehicles and the driving speed, the cost difference needing to be paid when the vehicle speed is adjusted and the lane change is adjusted is considered, the behavior selection adjustment coefficient is introduced, the ordering adjustment coefficient is added when the risk avoiding sequence is considered to be different, and then the cost source equation is
Figure RE-GDA0002364163650000051
Wherein k ismnRepresents EmSelecting an adjustment parameter, k, for the corresponding payment cost behavior when selecting the nth decision schememsqRepresents EmThe payment cost arrangement sequence corresponding to the adjustment sequence adjusts the parameter, k1Represents EmDistance dependent payment cost weight, k2Represents EmSpeed of rotationAssociated payment cost weight,/mRepresents EmLength from potential conflict point, vmRepresents EmThe current driving speed.
Determining the decision combination possibly generated by each vehicle in the step 9-5, wherein the number of the decision groups is n in total based on the cost function of n decision schemes corresponding to M vehiclesMAll arrangement forms thereof are Dcom＝(E2zn,E3zn,…,EMzn)(n＝1,2,…,N)。
Determining the payment cost required by each vehicle in each decision group in the step 9-6, determining the payment cost required by each vehicle in each decision group according to the arrangement result of all decision groups, wherein the decision group cost determination mode is
Figure BDA0002273851060000068
(zcom)＝f(Dcom)→g(zn) Wherein
Figure BDA0002273851060000069
(zcom) Means the cost combination corresponding to each decision group, f (D)com)→g(zn) Refers to the corresponding relationship between each decision group and each vehicle to the nth decision scheme cost function.
Establishing a vehicle driving intention interactive negotiation weight matrix in the step 9-7, determining the initial cost to be paid by each vehicle in each decision group, introducing a negotiation strategy theory into each vehicle behavior adjustment decision process, and enabling each vehicle E to be in a state of being matched with the initial costpFor the rest vehicles Eq(p ≠ q) specifies the weights for releasing
Figure BDA0002273851060000061
(p、q＝1,2,…,M)，
Figure BDA0002273851060000062
Reflects the preference degree of the current vehicle to other vehicle behavior decisions in the negotiation process, and
Figure BDA0002273851060000063
satisfy the need of
Figure BDA0002273851060000064
And 9-8, establishing a vehicle intention negotiation equation set according to the weight matrix, and solving a negotiation vector. From step 9-7, the vehicle intention negotiation weight matrix is constructed as
Figure BDA0002273851060000065
Then a vehicle intention negotiation equation set is established according to the weight matrix as
Figure BDA0002273851060000066
Solving the unique solution rho (rho) of the negotiation equation set1,ρ2,…,ρM) Namely the multi-vehicle conflict risk avoidance negotiation vector.
And 9-9, calculating the target total cost corresponding to each group of decision-making schemes, obtaining the minimum value of the total cost, and determining the scheme corresponding to the minimum value of the total cost. The target total cost function corresponding to each group of decision schemes is g*(zn)＝k3[ρ1g1(zn)+ρ2g2(zn)+…+ρMgM(zn)]WhereinAnd representing the safety coefficient of the corresponding scheme group reflected by the scheme composition structure in each decision group. By function g*(zn) And calculating the target total cost of each group, and comparing the calculation results, wherein the scheme group corresponding to the minimum value of the target total cost is the negotiation decision scheme group with higher acceptance degree of each vehicle.
Dfinal＝ming*(zn)→Dcom,DcomGreater than 1 group, depending on the strategy group number, smaller group as final decision, where DfinalRefers to a negotiation decision scheme group, g, corresponding to the minimum value of the target total cost*(zn)→DcomRefers to the process of finding the corresponding negotiation decision scheme set after the minimum value of the target assembly is determined.
The invention has the beneficial effects that: the decision-making suggestion is actually provided for the multi-vehicle danger avoiding and releasing process of the common bottleneck highway section, and the technical service support is provided for the further development of the self-adaptive cruise of the automatic driving vehicle.
Drawings
FIG. 1 is a flow chart of the overall logic provided by the present invention.
FIG. 2 is a flow chart of the present invention for updating the intensity sequence required for the same time by vehicle type.
Fig. 3 is a flowchart for updating the same time demand intensity sequence by interactive negotiation according to the vehicle driving intention.
Fig. 4 is a simulation experiment parameter setting provided by the present invention.
Fig. 5 is a trend graph of total cost required to be paid after interactive negotiation of each decision group provided by the present invention.
Fig. 6 is a comparison graph of cost saving rates of various scenes in a simulation experiment provided by the present invention.
Detailed Description
The following describes in detail a decision method for avoiding a risk of a collision of multiple vehicles on a frequently-occurring bottleneck road section of an expressway, which is provided by the invention, with reference to the accompanying drawings and specific examples. The embodiment is only used for explaining one situation in the technical solution of the present invention, and the embodiment is not used for limiting the scope of the present invention.
Embodiments of the present invention will be described below with reference to the drawings.
Fig. 1 shows a flowchart of an autonomous decision method for multi-vehicle collision risk avoidance on a frequently-occurring bottleneck section of a highway according to an embodiment of the present invention. The multi-vehicle collision avoidance autonomous decision method provided by the embodiment of the invention as shown in fig. 1 comprises the following steps:
firstly, the running state and the running environment of each vehicle are known and can be acquired by other vehicles in the running process, the real-time distance between the vehicles is included, parameters such as the real-time running speed of each vehicle are included, and high-quality information interaction can be realized between the vehicles. The present embodiment will be described in detail on the premise that the same time demand strength and the same vehicle type occur at most 2 times in the M-vehicle conflict state.
Step 1: and for a plurality of vehicles with conflicts, performing conflict resolution priority initial sequencing according to the time requirement strength by using a method that parallel numbers occupy the same sequencing space.
In the step, the multiple vehicles are arranged in a collision risk avoiding and releasing sequence according to the time demand intensity of the ith vehicle for automatic driving, which is represented by tau, and
Figure BDA0002273851060000081
(i ═ 1,2,3, …, M.M represents the number of vehicles in the current conflict state, l represents the length of autonomous vehicle i from the potential conflict point, and v represents the current travel speed of autonomous vehicle i). The time demand intensity priority level comparison is carried out on the vehicles with potential conflict according to tau, and the sequential numbering which is increased from 1 to 1 is carried out by adopting a method that parallel numbering occupies the same sequencing space.
According to the research of the existing related parameter values, the minimum headway of the automatic driving vehicle is 0.6s, the minimum headway of the manual driving vehicle is 1.3s, the whole transmission and reaction process is considered, the middle value 0.95s of the headway of the automatic driving vehicle and the manual driving vehicle is taken as a reference, and the comparison relation of the time demand strength is further described.
When in use
Figure BDA0002273851060000082
Defining that the time demand intensity of the two vehicles is the same; when in use
Figure BDA0002273851060000083
When is, i.e. taui＞τi-1Defining that the time demand intensity of the CAV vehicle with the number i is greater than the time demand intensity of the CAV vehicle with the number i-1, and determining that the CAV vehicle with the number i is sequenced before the CAV vehicle with the number i-1; when in use
Figure BDA0002273851060000084
When is, i.e. taui＜τi-1And when the time demand intensity of the CAV vehicle with the number i is smaller than the time demand intensity of the CAV vehicle with the number i-1, determining that the CAV vehicle with the number i-1 is ranked first with the CAV vehicle with the number i.
Taking M vehicles as an example, the M vehicles are ranked according to the time demand intensity and the conflict resolution initial sequence is set, and the mth vehicle is assumed to be1Vehicle and m1+1 vehicles of the same type with the same time demand intensity, m2And m is2The +1 vehicle time demand intensity and the vehicle type are the same. The result of the initial sequencing of M vehicles is [1,2. -. M ]1-m1-...-m2-m2-,M]。
Step 2: and judging whether the same time requirement intensity exists in the initial sequencing of the conflict resolution priority. The initial ranking results of the multiple vehicle time demand intensities will have two results of the same vehicle time demand intensity and the same vehicle time demand intensity.
If there is a vehicle time demand equal, the part of the initial sequence with the same vehicle time demand intensity number is extracted and transferred to step 3, where [ -m1-m1-]、[-m2-m2-]Extracting and transferring to the step 3.
If there are no vehicles with the same time demand level, the initial sequence is transferred to step 3.
And step 3: determining sequence [ -m ] with same switching-in time requirement intensity1-m1-]、[-m2-m2-]Corresponding vehicle, and [ -m ]1-m1-]、[-m2-m2-]And the corresponding vehicle goes to step 4; the number of the sequences of the remaining vehicle time demand intensities is kept unchanged, and the original sequence of the time demand intensities [1,21-m1-...-m2-m2-,M]Go to step 8.
And 4, step 4: comparing the vehicle types of the vehicles with the same time demand intensity, and carrying out the sequence [ -m ] with the same time demand intensity1-m1-]、[-m2-m2-]The temporary marking is carried out and the marking is carried out,is provided with
Figure BDA0002273851060000091
And checking whether each vehicle in the same time demand intensity sequence has the same vehicle type, if so, turning to the step 5, and if not, turning to the step 7.
Step 6: and determining the priority sequence updating values of the same vehicle types under the same time demand intensity according to the vehicle driving intention, and transferring the updating values to the step 8.
Fig. 2 shows a flowchart for updating the intensity sequence required for the same time according to the vehicle type, which includes the following specific processes:
step 6-1: and extracting the same vehicle type sequence under the same time demand intensity, and temporarily marking each vehicle under the condition. Then the initial set extraction sequence [ -m ] according to this embodiment2-m2-]And the corresponding vehicle are marked with
Figure BDA0002273851060000092
Step 6-2: and generating random unequal probabilities of the running adjustment intentions of the vehicles, and determining the corresponding relation between the random unequal probabilities and the vehicles. And quantifying the vehicle driving intention by using a random probability method, realizing the interaction process of the vehicle driving intention, only considering the adjustment intention intensity degree in the vehicle driving process, generating random unequal probabilities corresponding to the vehicles one by one under the conditions by using the random probability method, and determining the corresponding relation between the random unequal probabilities and the vehicles.
Step 6-3: and sequencing the random unequal probability of each vehicle from small to large on the basis of the determination of each vehicle and the corresponding random unequal probability, wherein the larger the probability value is, the stronger the contribution intention of the vehicle is. If the vehicle is
Figure BDA0002273851060000101
Corresponding random unequal probability greater than that of vehicle
Figure BDA0002273851060000102
Corresponding random unequal probability is obtainedArranged at the position of
Step 6-4: and correspondingly adjusting the vehicle arrangement position according to the random unequal probabilities.
Step 6-5: updating the current sequence according to random unequal probability, wherein the vehicle number corresponding to the minimum random probability is 0, the vehicle number corresponding to the next minimum random probability is 1, and the like is repeated until the numbering is finished, and then the sequence is updated
Figure BDA0002273851060000104
The updated value is determined as
Figure BDA0002273851060000105
And 7: and determining the updating value of the initial sequence of the demand intensity of the vehicles at the same time according to the priority of the vehicle types, and transferring the updating result to the step 8. According to the existing research, the minimum headway of an automatic driving vehicle is 0.6s, the minimum headway of a manual driving vehicle is 1.3s, a vehicle type priority sequence list is defined, a vehicle type priority sequence list is determined to be that the manual driving vehicle is larger than the automatic driving vehicle to be called by the embodiment, and then sequential numbering is carried out on a plurality of conflicting vehicles according to the vehicle type priority sequence, wherein the sequential numbering is carried out by utilizing a method that parallel numbering occupies the same sequencing space, the priority levels of the vehicle types are increased from high to low and are correspondingly increased from 0 to 1. The initial setting of the extraction sequence [ -m ] according to the present embodiment1-m1-]And corresponding vehicles, and making temporary marks
Figure BDA0002273851060000106
If the vehicle is
Figure BDA0002273851060000107
Flight performance priority lower than
Figure BDA0002273851060000108
Then pair
Figure BDA0002273851060000109
Is determined as
Figure BDA00022738510600001010
And 8: and (4) performing sequence updating again on the initial sequence of the remaining vehicle time demand intensity in the step (3) and the updated sequences transferred in the steps (6) and (7) to obtain a final sequence of the conflict risk avoidance and release priority, and transferring the sequence to the step (9).
According to the initial setting of this embodiment, the time demand intensity initial sequence is [1,2. -. m. -1-m1-...-m2-m2-,M]By using
Figure RE-GDA00023641636500001011
To pair
Figure RE-GDA00023641636500001012
The updating is carried out, and the updating is carried out,
Figure RE-GDA00023641636500001013
to [ -m ]2-m2-]Is updated by the process of
Figure RE-GDA0002364163650000111
The final sequence result is
Figure RE-GDA0002364163650000112
And step 9: and (3) on the basis of the final conflict risk avoidance and disengagement sequence, carrying out multi-vehicle behavior adjustment negotiation group decision by combining the vehicle driving intention, obtaining each vehicle behavior adjustment mode, and turning to the step 10.
Fig. 3 shows a flowchart for updating the intensity sequence of the same time requirement by interactive negotiation according to the vehicle driving intention, which specifically includes:
step 9-1: and determining the number corresponding to each vehicle according to the final sequence of the conflict resolution of the plurality of vehicles. The method is that m is equal to | Sort |, wherein m represents a vehicle number, and Sort represents a multi-vehicle conflict resolution sequenceAnd | | represents that the corresponding vehicle number is generated according to the conflict resolution ranking of the multiple vehicles. According to the initial setting of the present embodiment, the order number corresponding to each vehicle sudden-release sequence is
Figure BDA0002273851060000112
Step 9-2: a selectable adjustment scheme for each vehicle is determined. Using znThe N-th (N-1, 2, …, N) decision scheme adopted by the current vehicle is shown, and N-1 speed adjustment and N-2 lane change adjustment are selected as the decision schemes used in the present embodiment.
Step 9-3: constructing a cost function to be paid when each vehicle selects the same decision scheme, wherein the specific formula is g (z)n)＝G(g2(zn),g3(zn),…,gM(zn))。
Step 9-4: the payment cost source equation for definitely selecting a scheme for the vehicle is
Figure BDA0002273851060000113
Step 9-5: determining the decision combination possibly generated by each vehicle, and according to the initial setting of the embodiment, the number of the decision groups is n in total on the basis of the cost functions of n decision schemes corresponding to the M vehiclesMAll arrangement forms thereof are Dcom＝(E1zn,E2zn,…,EMzn)(n＝1,2,…,N)。
And 9-6: and determining the required payment cost of each vehicle in each decision group. And calculating the payment cost required by each vehicle in each decision group according to the cost source equation in the step 9-4.
Step 9-7: establishing a vehicle driving intention negotiation weight matrix of
Figure BDA0002273851060000114
And 9-8: establishing a pilot intention negotiation equation set according to the weight matrix as
Figure BDA0002273851060000121
Solving for a negotiation vector ρ ═ p (ρ ═ p)1,ρ2,…,ρM)。
Step 9-9: and calculating the target total cost corresponding to each group of decision schemes to obtain the minimum total cost, and determining the scheme corresponding to the minimum total cost. The target total cost function corresponding to each group of decision schemes is g*(zn)＝k3[ρ1g1(zn)+ρ2g2(zn)+…+ρMgM(zn)]And calculating the target total cost of each group, and comparing the calculation results, wherein the scheme group corresponding to the minimum value of the target total cost is the negotiation decision scheme group with higher acceptance degree of each vehicle. Dfinal＝ming*(zn)→Dcom,DcomGreater than 1 group, a smaller group is made as a final decision depending on the policy group number, where DfinalRefers to the group of negotiation decision schemes corresponding to the minimum value of the target total cost.
Step 9-10 transfers the cost minimum correspondence scheme to P9.
Step 10: and adjusting the running behavior of each vehicle. And each vehicle carries out behavior adjustment according to the driving intention negotiation result, avoids conflict points, ensures the running safety of each vehicle and meets the driving preference and benefit requirements of each vehicle as far as possible.
The main scenes in the simulation experiment process of the embodiment of the invention are set to be four, and the scenes are designed for the same time requirement intensity and the same vehicle type. And the second scene is designed for the same time demand intensity and different vehicle types. And the third scene is designed for different time demand intensity and the same vehicle type. And the fourth scene is designed for different time demand intensity and different vehicle types.
A conflict resolution behavior decision method is provided for a multi-vehicle conflict state, in order to reduce complexity and difficulty of an experimental process, parameter values such as running speed of each vehicle, distance from a potential conflict point and the like are set under certain conditions, a vehicle conflict resolution sequencing process is omitted, and multi-vehicle negotiation and behavior decision process simulation and analysis are directly performed on a given conflict resolution sequence.
The simulation process selects two typical vehicle types of automatic driving vehicles (CAV) and Manual Driving Vehicles (MDV) in common social vehicles as research objects, and the priority of the two vehicle types is defined as MDV & gtCAV.
Fig. 4 shows the setting of parameters such as vehicle type, driving speed, location, distance potential conflict point length, and the like in different scenes in the simulation experiment process according to the embodiment of the invention.
The driving intention negotiation weight matrix of each vehicle under each scene is generated as
Figure BDA0002273851060000131
Figure BDA0002273851060000132
Then, the negotiation vector is calculated as rho under each sceneScenario 1＝(0.259911,0.264718,0.216933,0.258438)， ρScenario 2＝(0.185993,0.232002,0.178007,0.241997,0.162001)，ρScenario 4= (0.188647,0.242559,0.255167,0.313627) and ρScenario 4＝(0.188647,0.242559,0.255167,0.313627)。
Fig. 5 shows the total cost required to be paid by each decision group obtained by negotiating the vector in each scene, so that the decision combination corresponding to the minimum payment cost in each scene is determined, and the avoidance adjustment mode of each vehicle is determined.
Fig. 6 shows the cost saving rate of each vehicle collision avoidance adjustment process realized by negotiation decision in each scene, which illustrates that the method of the present invention can not only realize each vehicle collision avoidance autonomous negotiation decision, but also reduce the total payment cost of the system, and is feasible.

Claims (10)

1. The cooperative fleet conflict risk avoidance autonomous decision-making method for the common bottleneck sections of the expressway is characterized by comprising the following steps of:
step 1: for a plurality of vehicles with conflict, initially arranging the conflict resolution sequence by using a method that parallel numbers occupy the same sequencing space according to the time demand intensity of each vehicle;
step 2: judging whether the initial arrangement of the conflict resolution sequence has the same time requirement strength, if the time requirement strengths are different, determining the initial sequence as a final sequence, and turning to the step 9; if the existing time of each vehicle is the same with the required intensity, the sequence is switched to the step 3;
and step 3: extracting vehicles corresponding to the same time requirement intensity in the initial sequence, and transferring the same time requirement intensity sequence to the step 4; the remaining vehicle time demand intensity sequence is kept unchanged, and the original time demand intensity initial sequence is transferred to step 8;
and 4, step 4: the vehicle type priority ranking is carried out on the extracted vehicles with the same time demand intensity, whether the same vehicle type exists in each vehicle under the same time demand intensity is judged, if the same vehicle type exists in each vehicle under the same time demand intensity, the step 5 is carried out, and if the same vehicle type does not exist in each vehicle under the same time demand intensity, the step 7 is carried out;
and 5: extracting the same vehicle type sequence under the same time demand intensity, and transferring the sequence to the step 6;
step 6: performing intention interaction on vehicles of the same vehicle type under the same time requirement, renumbering the initial sequences of the vehicles according to the driving intention of the vehicles, and transferring the updating result to the step 8;
and 7: updating the initial sequence of the required intensity of the vehicles at the same time according to different types of the vehicles, and transferring the updating result to the step 8;
and 8: performing sequence updating again on the initial sequence of the remaining vehicle time demand intensity in the step 3 and the updated sequences converted in the steps 6 and 7 to obtain a final sequence of conflict resolution, and converting the sequence to a step 9;
and step 9: on the basis of the final sequence of conflict resolution, carrying out multi-vehicle group negotiation decision by combining vehicle intentions, determining each vehicle behavior adjustment mode, and turning to the step 10;
step 10: and adjusting the running behavior of each vehicle.
2. The autonomous decision-making method for collision avoidance of cooperative fleet bottleneck sections on expressway according to claim 1, wherein: the method for occupying the same sorting space by utilizing parallel numbers for a plurality of conflicting vehicles carries out initial sorting of conflict resolution priority according to the time demand intensity of the vehicles, and the time demand intensity of the ith vehicle for automatic driving is expressed as tau, andwherein i-1, 2,3, …, M.M represents the number of vehicles in the current conflict state, l represents the length of the autonomous vehicle i from the potential conflict point, and v represents the current running speed of the autonomous vehicle i; and comparing the time demand intensity priority levels of the automatic driving vehicles with potential conflicts according to tau, and numbering the automatic driving vehicles in an ascending order from 1 to 1 in a way that parallel numbers occupy the same sequencing space.
3. The autonomous decision-making method for collision avoidance of cooperative fleet bottleneck sections on expressway according to claim 1, wherein: determining the adjustment mode of each vehicle driving behavior, and performing multi-machine negotiation group decision of the flight behavior by combining the vehicle flight intention on the basis of the final conflict resolution priority sequence, wherein the method comprises the following steps:
step 9-1: determining the number corresponding to each vehicle according to the final sequence of the conflict resolution of the plurality of vehicles;
step 9-2: determining a selectable adjustment scheme for each vehicle;
step 9-3: constructing cost functions to be paid when each vehicle selects the same decision scheme;
step 9-4: defining a payment cost source equation of each vehicle for selecting a scheme;
step 9-5: determining a decision combination which can be generated by each vehicle;
and 9-6: determining the payment cost required by each vehicle in each decision group;
step 9-7: establishing a vehicle driving intention interactive negotiation weight matrix;
and 9-8: establishing a vehicle driving intention negotiation equation set according to the weight matrix, and solving a negotiation vector;
step 9-9: calculating the target total cost corresponding to each group of decision-making schemes to obtain the minimum total cost, and determining the scheme corresponding to the minimum total cost;
step 9-10: the cost minimum correspondence scheme is transferred to step 10.
4. The method of claim 3, wherein the method comprises the following steps: determining the corresponding serial numbers of all vehicles on the basis of the final sorting of the conflict resolution of the multiple vehicles, wherein the method for determining the corresponding serial numbers of all vehicles according to the final sorting of the conflict resolution of the multiple vehicles is m ═ Sort |, wherein m represents the serial numbers of the vehicles, Sort represents the conflict resolution sequence of the multiple vehicles, and | | represents the generation of the corresponding serial numbers of the vehicles according to the rank of the conflict resolution of the multiple vehicles; by using EmThis indicates a vehicle in which the M-th (M-1, 2,3, …, M) vehicle makes a collision resolution decision.
5. The method of claim 3, wherein the method comprises the following steps: determining vehicle-selectable adjustments using znRepresents the N (N-1, 2, …, N) th decision scheme adopted by the current vehicle.
6. The method of claim 3, wherein the method comprises the following steps: constructing a cost function to be paid when each vehicle selects the same decision scheme, wherein the specific cost function is g (z)n)＝G(g1(zn),g2(zn),…,gM(zn) Wherein z isnIndicates the nth decision scheme, g (z), taken by the vehiclen) Representing the cost function, g, of each vehicle to the nth decision schemem(zn) Represents EmInitial cost for the nth decision scheme, G (G)1(zn),g2(zn),…,gM(zn) Is) represents the comparison of the initial cost of the nth decision scheme for each member in the system.
7. The method of claim 3, wherein the method comprises the following steps: determining a payment cost source equation of a certain scheme selected by a vehicle, introducing a behavior selection adjustment coefficient according to the position relation and the flying speed between two vehicles and considering the cost difference required to be paid when the vehicle speed is adjusted and the lane change is adjusted, and adding a sequencing adjustment coefficient according to different releasing sequences, wherein the cost source equation isWherein k ismnRepresents EmSelecting an adjustment parameter, k, for the corresponding payment cost behavior when selecting the nth decision schememsqRepresents EmThe payment cost arrangement sequence corresponding to the adjustment sequence adjusts the parameter, k1Represents EmDistance-associated payment cost weight, k2Represents EmVelocity associated payment cost weight,/mRepresents EmLength from potential conflict point, vmRepresents EmThe current driving speed.
8. The method of claim 3, wherein the method comprises the following steps: determining the possible decision combination of each vehicle, and counting n decision groups based on the cost function of n decision schemes corresponding to M vehiclesMAll arrangement forms thereof are Dcom＝(E1zn,E2zn,…,EMzn)(n＝1,2,…,N)。
9. The method of claim 3, wherein the method comprises the following steps: determining the requirements of each vehicle in each decision groupPaying cost, determining the cost required to be paid by each vehicle in each decision group according to the arrangement results of all decision groups, wherein the decision group cost determination mode is
Figure FDA0002273851050000041
Wherein
Figure FDA0002273851050000042
Means the cost combination corresponding to each decision group, f (D)com)→g(zn) The corresponding relation between each decision group and each vehicle to the nth decision scheme cost function is indicated.
10. The highway frequent bottleneck road section cooperation fleet conflict hedging autonomous decision method according to claim 3, characterized in that: establishing a vehicle intention interactive negotiation weight matrix, determining the initial cost to be paid by each vehicle in each decision group, introducing a negotiation strategy theory into the decision process of each vehicle, and enabling each vehicle EpFor the rest vehicles Eq(p ≠ q) specifies the weights for releasing
Figure FDA0002273851050000044
The magnitude of (c) represents the preference of the current vehicle for the rest of the vehicle behavior decisions during the negotiation,
Figure FDA0002273851050000045
satisfy the requirement of
Establishing a vehicle driving intention negotiation equation set according to the weight matrix, solving a negotiation vector, and establishing a vehicle negotiation weight matrix of
Figure FDA0002273851050000047
Then the vehicle driving intention is established according to the weight matrixThe system of negotiation equations is
Figure FDA0002273851050000048
Solving the unique solution rho (rho) of the negotiation equation set1,ρ2,…,ρM) I.e. a negotiation vector for resolving multiple vehicle conflicts.
Calculating the target total cost corresponding to each group of decision schemes to obtain the minimum total cost, determining the scheme corresponding to the minimum total cost, wherein the target total cost function corresponding to each group of decision schemes is g*(zn)＝k3[ρ1g1(zn)+ρ2g2(zn)+…+ρMgM(zn)]Wherein
Figure FDA0002273851050000049
Representing the safety coefficient of the corresponding scheme group reflected by the scheme composition structure in each decision group through a function g*(zn) Calculating the target total cost of each group, and comparing the calculation results, wherein the scheme group corresponding to the minimum value of the target total cost is the negotiation decision scheme group with higher acceptance degree of each vehicle; dfinal＝ming*(zn)→Dcom,DcomGreater than 1 group, a smaller group is determined as a final decision depending on the strategy group number, where DfinalRefers to a negotiation decision scheme group, g, corresponding to the minimum value of the target total cost*(zn)→DcomThe method refers to a process of searching a corresponding negotiation decision scheme set after the minimum value of the target total cost is determined.
CN201911115220.8A 2019-11-14 2019-11-14 Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway Active CN110853335B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911115220.8A CN110853335B (en) 2019-11-14 2019-11-14 Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911115220.8A CN110853335B (en) 2019-11-14 2019-11-14 Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway

Publications (2)

Publication Number Publication Date
CN110853335A true CN110853335A (en) 2020-02-28
CN110853335B CN110853335B (en) 2020-11-27

Family

ID=69601794

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911115220.8A Active CN110853335B (en) 2019-11-14 2019-11-14 Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway

Country Status (1)

Country Link
CN (1) CN110853335B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112652189A (en) * 2020-12-30 2021-04-13 西南交通大学 Traffic distribution method, device and equipment based on policy flow and readable storage medium
CN113593228A (en) * 2021-07-26 2021-11-02 广东工业大学 Automatic driving cooperative control method for bottleneck area of expressway

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101859494A (en) * 2009-04-06 2010-10-13 通用汽车环球科技运作公司 Autonomous vehicle management
CN102314769A (en) * 2010-07-07 2012-01-11 交通运输部公路科学研究所 Intelligent safety early-warning control method for merging area of entrance ramp of expressway
US20130116915A1 (en) * 2010-07-16 2013-05-09 Universidade Do Porto Methods and Systems For Coordinating Vehicular Traffic Using In-Vehicle Virtual Traffic Control Signals Enabled By Vehicle-To-Vehicle Communications
CN103700251A (en) * 2013-11-27 2014-04-02 东南大学 Variable speed limiting and ramp control coordination and optimization control method on expressway
CN103733237A (en) * 2011-07-05 2014-04-16 高通股份有限公司 Road-traffic-based group, identifier, and resource selection in vehicular peer-to-peer networks
CN105931460A (en) * 2016-05-13 2016-09-07 东南大学 Variable speed limit control strategy optimization method for continuous bottleneck section of expressway
CN106781435A (en) * 2016-12-15 2017-05-31 清华大学 A kind of Fei Xinkong intersections platooning passing method based on radio communication
CN107036600A (en) * 2015-09-30 2017-08-11 小蚁科技（香港）有限公司 System and method for autonomous vehicle navigation
CN107085978A (en) * 2017-06-21 2017-08-22 南京航空航天大学 A kind of control aid decision instruction generation method based on required arrival time
CN107430807A (en) * 2015-03-31 2017-12-01 爱信艾达株式会社 Automatic Pilot accessory system, automatic Pilot householder method and computer program
CN107516439A (en) * 2017-07-28 2017-12-26 北京航空航天大学 It is a kind of based on the aircraft conflict Resolution method and device for being satisfied with game theory

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101859494A (en) * 2009-04-06 2010-10-13 通用汽车环球科技运作公司 Autonomous vehicle management
CN102314769A (en) * 2010-07-07 2012-01-11 交通运输部公路科学研究所 Intelligent safety early-warning control method for merging area of entrance ramp of expressway
US20130116915A1 (en) * 2010-07-16 2013-05-09 Universidade Do Porto Methods and Systems For Coordinating Vehicular Traffic Using In-Vehicle Virtual Traffic Control Signals Enabled By Vehicle-To-Vehicle Communications
CN103733237A (en) * 2011-07-05 2014-04-16 高通股份有限公司 Road-traffic-based group, identifier, and resource selection in vehicular peer-to-peer networks
CN103700251A (en) * 2013-11-27 2014-04-02 东南大学 Variable speed limiting and ramp control coordination and optimization control method on expressway
CN107430807A (en) * 2015-03-31 2017-12-01 爱信艾达株式会社 Automatic Pilot accessory system, automatic Pilot householder method and computer program
CN107036600A (en) * 2015-09-30 2017-08-11 小蚁科技（香港）有限公司 System and method for autonomous vehicle navigation
CN105931460A (en) * 2016-05-13 2016-09-07 东南大学 Variable speed limit control strategy optimization method for continuous bottleneck section of expressway
CN106781435A (en) * 2016-12-15 2017-05-31 清华大学 A kind of Fei Xinkong intersections platooning passing method based on radio communication
CN107085978A (en) * 2017-06-21 2017-08-22 南京航空航天大学 A kind of control aid decision instruction generation method based on required arrival time
CN107516439A (en) * 2017-07-28 2017-12-26 北京航空航天大学 It is a kind of based on the aircraft conflict Resolution method and device for being satisfied with game theory

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
MIN YANG ET AL.: ""Exploring the impact or residential relocation on modal shift in commute trips Evidence from a quasi-longitudinal analysis"", 《TRANSPORT POLICY》 *
卢志平: ""基于过程交互的复杂动态群体决策方法研究"", 《中国博士学位论文库 经济与管理科学辑》 *
江欣国 等: ""基于元胞自动机的高速公路临时瓶颈交通流仿真"", 《西南交通大学学报》 *
王硕 等: ""低空救援航空器航迹预测与冲突解脱研究"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 *
盘意伟: "" 快速路入口匝道合流区的瓶颈形成机理与特性研究"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112652189A (en) * 2020-12-30 2021-04-13 西南交通大学 Traffic distribution method, device and equipment based on policy flow and readable storage medium
CN112652189B (en) * 2020-12-30 2021-09-28 西南交通大学 Traffic distribution method, device and equipment based on policy flow and readable storage medium
CN113593228A (en) * 2021-07-26 2021-11-02 广东工业大学 Automatic driving cooperative control method for bottleneck area of expressway

Also Published As

Publication number Publication date
CN110853335B (en) 2020-11-27

Similar Documents

Publication Publication Date Title
CN111898211B (en) Intelligent vehicle speed decision method based on deep reinforcement learning and simulation method thereof
CN111768637B (en) Signal intersection traffic signal lamp and vehicle track control method
JP6931937B2 (en) A learning method and learning device that uses human driving data as training data to perform customized route planning by supporting reinforcement learning.
CN110853335B (en) Cooperative fleet conflict risk avoidance autonomous decision-making method for common bottleneck sections of expressway
CN106874597B (en) highway overtaking behavior decision method applied to automatic driving vehicle
CN106672029B (en) Control the method and device of train operation
CN108877256B (en) Wireless communication-based method for controlling scattered cooperative self-adaptive cruise near intersection
CN111332283A (en) Method and system for controlling a motor vehicle
CN105843207B (en) Vehicle-state forecasting system
Qian et al. Autonomous intersection management systems: Criteria, implementation and evaluation
JP2014211714A (en) Rank traveling system
CN113489793B (en) Expressway double-lane cooperative control method in mixed traffic scene
CN111775938B (en) Lane change path planning method, device and system
EP3705367B1 (en) Training a generator unit and a discriminator unit for collision-aware trajectory prediction
CN106251012A (en) The path calculation method of a kind of band weak rock mass logistics transportation scheduling and device
CN113791615A (en) Hybrid vehicle queue distributed model prediction control method
KR20220014284A (en) Gear-shift control data generation method, gear-shift control device, and gear-shift control system
CN117252318B (en) Intelligent networking automobile group machine collaborative carpooling scheduling method and system
CN113344336A (en) Vehicle scheduling method and device and storage medium
DE102021113042A1 (en) ALGORITHM FOR RESOLVING A ROAD RESOURCE CONFLICT
Guo et al. Modeling, learning and prediction of longitudinal behaviors of human-driven vehicles by incorporating internal human DecisionMaking process using inverse model predictive control
CN110641470A (en) Pure electric vehicle driving auxiliary system optimization method integrating driver preference
CN114475607A (en) Method and device for changing lane for automatic driving vehicle, vehicle and storage medium
Yan et al. Application of fuzzy control under time-varying universe in unmanned vehicles
CN110954341B (en) Test method and system for intelligent networking automobile test scene

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant