CN-121982917-A - CAV (computer aided design) special road layout optimization method based on travel time reliability

CN121982917ACN 121982917 ACN121982917 ACN 121982917ACN-121982917-A

Abstract

The invention discloses a CAV special road layout optimization method based on travel time reliability, which comprises the steps of constructing an expanded road network comprising a conventional road section and a virtual road section, calculating the traffic capacity of the conventional road section and the virtual road section based on the expanded road network according to a vehicle following mode and a corresponding safety head time interval, establishing a mixed traffic flow distribution model based on the traffic capacity, solving the mixed traffic flow distribution model to obtain road network balanced flow distribution, acquiring travel time sample sets of the road section and a path through Monte Carlo random sampling according to the balanced flow distribution, calculating road network buffer time indexes based on the travel time sample sets, wherein the buffer time indexes represent the road network travel time reliability, and establishing a CAV special road layout optimization model by taking the layout quantity of CAV special roads on the virtual road section as decision variables and taking the minimum travel time of the road network and the minimum road network total buffer time index as optimization targets, wherein the model is used for outputting a CAV special road layout optimization scheme.

Inventors

LI TONGFEI
ZHAO YARU
WENG JIANCHENG
DOU XUEPING
FAN BO
ZHANG YONGNAN

Assignees

北京工业大学

Dates

Publication Date: 20260505
Application Date: 20260206

Claims (9)

1. The CAV special lane layout optimization method based on travel time reliability is characterized by comprising the following steps of: Constructing an extended road network comprising a conventional road section and a virtual road section; Based on the expanded road network, calculating the traffic capacities of the conventional road section and the virtual road section according to the vehicle following mode and the corresponding safe headway; Establishing a mixed traffic flow distribution model based on the traffic capacity, and solving the mixed traffic flow distribution model to obtain road network balanced flow distribution; Acquiring travel time sample sets of road sections and paths through Monte Carlo random sampling according to the balanced flow distribution; Calculating a road network buffer time index based on the travel time sample set, wherein the buffer time index characterizes the travel time reliability of the road network; And taking the layout quantity of CAV special lanes on the virtual road section as a decision variable and taking the minimum total travel time of the road network and the minimum total buffer time index of the road network as optimization targets to establish a CAV special lane layout optimization model, wherein the CAV special lane layout optimization model is used for outputting a set of CAV special lane layout optimization schemes.
2. The CAV private road layout optimizing method based on travel time reliability as recited in claim 1, wherein constructing the expanded road network comprises: and adding a virtual road section to each road section of the initial road network, marking the initial road section as a conventional road section, wherein the sum of the lane numbers of the conventional road section and the virtual road section is equal to the total number of lanes of the corresponding road section in the initial road network, the conventional road section allows the online automatic driving vehicle to mix with the human driving vehicle, and the virtual road section only allows the online automatic driving vehicle to pass through.
3. The CAV lane layout optimization method based on travel time reliability as recited in claim 1, wherein the vehicle following pattern comprises: The conventional road sections are divided into five modes of human-driven vehicles following human-driven vehicles, human-driven vehicles following net-connected automatic driving vehicles, net-connected automatic driving vehicles following human-driven vehicles, net-connected automatic driving vehicles in the same formation and net-connected automatic driving vehicles following net-connected automatic driving vehicles of different formations; The virtual road section is divided into two modes that the online automatic driving vehicles follow the online automatic driving vehicles in the same formation, and the online automatic driving vehicles follow the online automatic driving vehicles of different formations.
4. The CAV private road layout optimizing method based on travel time reliability as recited in claim 1, wherein the traffic capacity of the regular road section is: ; Wherein, the Is the traffic capacity of the conventional road section, For the number of lanes on a regular road segment, For the average headway on a regular road segment, 、、、、 For different vehicle following modes of the regular road section, 、、、 Is that 、、、、 Corresponding to different safe time intervals of the vehicle head, To expand the collection of conventional road segments in the road network; The traffic capacity of the virtual road section is as follows: ; Wherein, the As the traffic capacity of the virtual road section, For the average headway on the virtual road segment, 、 For different vehicle following modes of the virtual road segments, 、 Is that 、 Corresponding to different safe time intervals of the vehicle head, For the number of lanes on the virtual road segment, To expand the set of virtual road segments in the road network.
5. The CAV lane layout optimization method based on travel time reliability as recited in claim 1, wherein establishing a mixed traffic flow distribution model based on the traffic capacity, solving the mixed traffic flow distribution model to obtain road network balanced flow distribution comprises: constructing a variation inequality problem comprising a path cost function, a flow conservation constraint and a variable non-negative constraint based on the traffic capacity, wherein the path cost function is formed by aggregating road section travel time, and the road section travel time is calculated according to the road section flow and the traffic capacity through a BPR function; And solving the variation inequality problem by adopting an iterative algorithm based on path switching until a convergence condition is met, so as to obtain the balanced flow distribution.
6. The CAV lane layout optimization method based on travel time reliability as recited in claim 1, wherein obtaining travel time sample sets of road segments and paths through monte carlo random sampling according to the balanced flow distribution comprises: s1, generating a random arrangement sequence of the online automatic driving vehicle and the human driving vehicle on each road section based on the balanced flow distribution; s2, recognizing a following mode between front vehicles and rear vehicles in the random arrangement sequence, endowing corresponding safe headway, and calculating a road section average headway according to the safe headway of all vehicle pairs; s3, calculating the traffic capacity of the road section based on the average headway of the road section, and calculating the travel time of the road section by using a BPR function; s4, calculating the travel time of the path according to the road sections contained in the path and the travel time of the road sections; s5 repeating S1-S4 Next, each strip is obtained And constructing a path travel time set by using different travel time samples.
7. The CAV lane layout optimization method based on travel time reliability as recited in claim 1, wherein calculating a road network buffer time index based on the travel time sample set comprises: Extracting the 95 th quantile from the path travel time sample set to be used as planning travel time, and calculating the average value of the path travel time sample set to be used as average travel time; Dividing the difference between the planned travel time and the average travel time by the average travel time to obtain a path buffer time index; Performing flow weighted average on all path buffer time indexes between origin and destination pairs of the road network to obtain an origin and destination pair buffer time index; And carrying out flow weighted average on the buffer time indexes at all the origin and destination points to obtain the road network total buffer time index.
8. The CAV lane layout optimization method based on travel time reliability according to claim 1, wherein the constraint conditions of the CAV lane layout optimization model include lane resource conservation constraint, traffic conservation constraint, user balance constraint and generalized travel cost constraint.
9. The method for optimizing the CAV lane layout based on travel time reliability according to claim 1, wherein the CAV lane layout optimization model is used for outputting a set of CAV lane layout optimization schemes, and the CAV lane layout optimization model is solved by adopting a fast elite multi-objective genetic algorithm; The method for solving the CAV lane layout optimization model by adopting the fast elite multi-objective genetic algorithm comprises the following steps: s1, carrying out integer coding on a virtual road section layout scheme, and randomly generating an initial population; S2, generating a child population through a selection operation based on non-dominant sorting and crowding degree distance, a simulation binary crossover operation and a polynomial variation operation; S3, merging the parent population and the offspring population, then carrying out rapid non-dominant sorting, and screening individuals according to the sorting level and the crowding degree distance to form a new-generation population; S4, iteratively executing S1-S3 until the maximum evolution algebra is reached, and outputting the scheme in the non-dominant hierarchy as the set of CAV special channel layout optimization schemes.

Description

CAV (computer aided design) special road layout optimization method based on travel time reliability Technical Field The invention belongs to the technical field of intelligent Internet of things, and particularly relates to a CAV special road layout optimization method based on travel time reliability. Background With the rapid development of intelligent networking technology, a typical traffic scenario is that networked autonomous vehicles (Connected and Autonomous Vehicle, CAV) and Human-driven vehicles (Human-DRIVEN VEHICLE, HV) coexist and travel in a traffic road network. The safe headway corresponding to different vehicle combinations (such as HV following HV, CAV following CAV and the like) in the mixed running scene is different, and the arrangement sequence of vehicles on road sections is completely random, so that the road section average headway in the mixed running scene is a random variable related to the spatial distribution of different types of vehicles on the road sections, and the road section average headway directly determines the road section traffic capacity, and therefore the road section traffic capacity in the mixed running scene is also a random variable. The size of the road section traffic capacity directly influences the travel time of the vehicle on the road section, so that under a mixed traffic scene, the road section travel time shows huge fluctuation due to the randomness of the road section traffic capacity, and the change of the road section travel time reliability is caused to be further conducted to the whole road network, and finally the travel time reliability of the whole road network is influenced, but a method capable of accurately quantitatively evaluating the road network travel time reliability under the mixed traffic scene is lacking currently. In addition, from the perspective of traffic planning and a manager, how to improve the travel time reliability of a road network through the optimal layout of CAV is also a great difficulty, so that the invention provides a CAV special lane layout method capable of considering the minimum total travel time of the road network and the optimal total travel time reliability of the road network, and can provide a scientific CAV special lane layout decision based on different travel efficiency requirements and different travel time reliability preferences for traffic managers. Disclosure of Invention In order to solve the technical problems, the invention provides a CAV special road layout optimization method based on travel time reliability, which can provide a traffic manager with a scientific CAV special road layout decision based on different travel efficiency requirements and different travel time reliability preferences. In order to achieve the above purpose, the present invention provides a CAV lane layout optimization method based on travel time reliability, comprising: Constructing an extended road network comprising a conventional road section and a virtual road section; Based on the expanded road network, calculating the traffic capacities of the conventional road section and the virtual road section according to the vehicle following mode and the corresponding safe headway; Establishing a mixed traffic flow distribution model based on the traffic capacity, and solving the mixed traffic flow distribution model to obtain road network balanced flow distribution; Acquiring travel time sample sets of road sections and paths through Monte Carlo random sampling according to the balanced flow distribution; Calculating a road network buffer time index based on the travel time sample set, wherein the buffer time index characterizes the travel time reliability of the road network; and taking the layout number of the CAV special lanes on the virtual road section as a decision variable, and taking the minimum total travel time of the road network and the minimum total buffer time index of the road network as optimization targets, establishing a CAV special lane layout optimization model, wherein the CAV special lane layout optimization model is used for outputting a set of CAV special lane layout optimization schemes. Optionally, constructing the extended road network includes: and adding a virtual road section to each road section of the initial road network, marking the initial road section as a conventional road section, wherein the sum of the lane numbers of the conventional road section and the virtual road section is equal to the total number of lanes of the corresponding road section in the initial road network, the conventional road section allows the online automatic driving vehicle to mix with the human driving vehicle, and the virtual road section only allows the online automatic driving vehicle to pass through. Optionally, the vehicle following mode includes: The conventional road sections are divided into five modes of human-driven vehicles following human-driven vehicles, human-driven vehicles following net-con