CAV multi-vehicle cooperative collision avoidance method in mixed flow environment based on composite network

Abstract

The present invention relates to the field of intelligent connected vehicle collision avoidance technology, and specifically refers to a method, device and computer-readable storage medium for collaborative collision avoidance of multiple intelligent connected vehicles in a mixed flow environment based on a composite network, including: establishing a composite network composed of intelligent connected vehicle state nodes and connection arcs, and finding a safe and feasible operation set according to the driving state of the intelligent connected vehicle; converting the collision avoidance problem into a network flow optimization problem, and estimating the optimal decision of multiple vehicles by finding the optimal network; realizing active lane changing and collision avoidance by dynamically iterating the speed and position of multiple vehicles; and performing simulation through SUMO software. The present invention proposes a new collision avoidance model with a wider application range; making full use of CACC technology to avoid traffic congestion; constructing a virtual composite network based on real traffic flow, optimizing the composite network, improving road traffic efficiency, and improving the traffic environment; integrating network model construction, optimal decision model and simulation verification into one, and fully verifying the effectiveness and feasibility of the model.

Description

Technical Field

[0001] The present invention relates to the field of vehicle collision avoidance technology, and in particular to a method, device and computer-readable storage medium for collaborative collision avoidance of multiple intelligent connected vehicles in a mixed flow environment based on a composite network. Background Art

[0002] With the development of economy and society, the level of urban modernization has been continuously improved, and intelligent transportation technology has been rapidly developed. At the same time, the complex and changeable urban road conditions, the increasingly serious traffic congestion, and the severe traffic safety situation have put forward new requirements for intelligent transportation systems. In the field of intelligent transportation systems, it is a very important research topic to efficiently use intelligent networked vehicle driving technology to achieve more coordinated and close cooperation between people, vehicles and roads. At present, based on the rapid development and popularization of intelligent transportation systems, intelligent networked vehicles (Connected-Automated Vehicle, CAV) and human-driven vehicles (Human-driven Vehicle, HV) sharing the road has become a foreseeable future, and the research on fleet cooperative driving systems is also gradually unfolding. Most of this research is based on two classic theories, namely the Intelligent Vehicle-Highway Systems Architecture and the Hybrid Dynamic System theory. With the continuous upgrading of Internet of Vehicles technology, the popularity of vehicle sensing, vehicle-to-vehicle communication and other technologies has been increasing. The active collision avoidance technology of networked vehicles, one of the core technologies of intelligent networked vehicles, is receiving more and more attention.

[0003] In the prior art, the road traffic research team of the University of California, Berkeley in the United States creatively proposed an intelligent transportation system with a hierarchical structure, which specifically includes a perception layer, a link layer, a coordination layer, a control layer and an actuator. The system preliminarily outlines the prototype of modern vehicle networking technology, but many details have not yet been perfected; Hoc JH and other scholars from the French National Academy of Sciences proposed that collaborative lane changing refers to when two or more multi-intelligent agents perform travel tasks, they coordinate and control through a cloud platform to alleviate resource tension or conflict and achieve individual or collective operation goals. The theory clarifies the information interaction behavior between vehicles and between vehicles and cloud platforms, but has not yet proposed a clear solution to traffic safety issues; Heesen M and other scholars from the German Aerospace Center set up a typical test scenario and modeled and analyzed the collaborative behavior in the lane changing process based on a logistic regression algorithm, but the scope of application is small and it is difficult to apply to other traffic scenarios such as ramps and intersections.

[0004] At present, the existing technology for vehicle collision avoidance has the following defects:

[0005] (1) All current collision avoidance models are logical selection models. However, the required logical selection models are not exactly the same in different traffic scenarios such as signalized intersections, three-lane, four-lane, and five-lane intersections. Designing different logical selection models for different situations has limitations for the actual research of vehicle collision avoidance technology.

[0006] (2) Current technical research is relatively single-minded and does not conform to the actual situation, which is not conducive to solving the problems of traffic congestion, traffic accidents and resource shortages in actual situations;

[0007] (3) The current collision avoidance method is that the vehicles involved perform collision avoidance maneuvers, which is difficult to solve the problem of traffic congestion caused by HV traffic accidents. Summary of the invention

[0008] To this end, the technical problem to be solved by the present invention is to overcome the need to design different logical selection collision avoidance models for different application scenarios in the prior art, which has limitations in research under different scenarios; a single technical research does not conform to the actual situation; and only performing obstacle avoidance operations on the vehicles involved is difficult to solve the problem of traffic congestion caused by HV traffic accidents.

[0009] In order to solve the above technical problems, the present invention provides a multi-vehicle collaborative collision avoidance method for intelligent connected vehicles in a mixed flow environment based on a composite network, comprising:

[0010] When a faulty vehicle appears on a lane during the driving of an intelligent connected vehicle and a human-driven vehicle, the next driving operation of each intelligent connected vehicle is set for all the intelligent connected vehicles that are normally driving on the lanes within a preset range; based on the collaborative adaptive cruise control system and the safety distance model, it is identified whether the next driving operation of each intelligent connected vehicle is feasible, and the next feasible driving operation of each intelligent connected vehicle is obtained;

[0011] The positions of all the intelligent connected vehicles that are normally traveling on the lanes within the preset range are marked respectively, and recorded as the initial state nodes of the corresponding intelligent connected vehicles; according to the feasible driving operations of each intelligent connected vehicle at the next moment, the virtual vehicle positions of each intelligent connected vehicle are obtained, and marked as the virtual vehicle state nodes of each intelligent connected vehicle; according to the initial state nodes of each intelligent connected vehicle and the corresponding virtual vehicle state nodes, the node sets of each intelligent connected vehicle are obtained, and the operation sequence sets of each intelligent connected vehicle are obtained;

[0012] Use connection arcs to represent the connection between each node in each intelligent connected vehicle node set and each node in the remaining intelligent connected vehicle node sets, and establish a composite network of connected vehicle collision avoidance behaviors based on all intelligent connected vehicle state nodes and connection arcs;

[0013] Based on the collaborative adaptive cruise control system, for any intelligent networked vehicle, the acceleration of the intelligent networked vehicle in the current lane and the acceleration of other intelligent networked vehicles behind the intelligent networked vehicle before the intelligent networked vehicle performs a feasible driving operation is obtained; the acceleration of the intelligent networked vehicle in the target lane and the acceleration of other intelligent networked vehicles behind the intelligent networked vehicle are obtained after the intelligent networked vehicle performs a feasible driving operation; the courtesy factor and the benefit factor are set, and according to each acceleration obtained, the attribute of the connection arc is defined to represent the negative benefit between two nodes, so as to obtain the value of different connection arcs of each intelligent networked vehicle;

[0014] In the composite network, a decision function is set, and the objective function of the target intelligent networked vehicle is established by combining the connection arcs; at the same time, to ensure that when the current node and the next node of the target intelligent networked vehicle are selected, both the current node and the next node are selected, constraint conditions are set; and it is stipulated that the sum of all decision functions of the target intelligent networked vehicle is 1 each time a connection arc is selected; according to all the connection arcs and nodes in the composite network, different driving routes of the target intelligent networked vehicle when iterating until a preset number of iterations is reached are obtained, and according to the value of each connection arc in each driving route, the value of the sum of all the connection arcs in each driving route is calculated, and by comparing the values ​​of the sum of all the connection arcs, the objective function is solved, and the driving route corresponding to the minimum sum of all the connection arcs of the target intelligent networked vehicle when iterating until the preset number of iterations is obtained, and the driving route is the optimal driving route of the target intelligent networked vehicle;

[0015] Based on the process of obtaining the optimal driving route of the above-mentioned target intelligent connected vehicle, the optimal driving routes of all intelligent connected vehicles are obtained in turn, so that all intelligent connected vehicles driving normally on the road behind the faulty intelligent connected vehicle can change lanes in advance according to their corresponding optimal driving routes to avoid the influence range of the preset faulty intelligent connected vehicle.

[0016] Preferably, the step of setting the driving operation of each intelligent networked vehicle at the next moment for all intelligent networked vehicles driving normally on lanes within a preset range respectively; based on the cooperative adaptive cruise control system and according to the safety distance model, identifying whether the driving operation of each intelligent networked vehicle at the next moment is feasible, and obtaining the feasible driving operation of each intelligent networked vehicle at the next moment includes:

[0017] According to the preset driving step length and driving direction, the next moment driving operation of each intelligent networked vehicle is set for all the intelligent networked vehicles that are normally driving on the lanes within the preset range; wherein the preset driving step length is +1 or -1 or +2; the preset driving direction is forward, backward, left, and right;

[0018] Based on the cooperative adaptive cruise control system, for the target intelligent connected vehicle Z i, i = 1, 2, 3..., m, the longitudinal distance between the target intelligent networked vehicle bumper and the bumper of the intelligent networked vehicle immediately ahead after the driving operation is recorded as D; the safety distance between the target intelligent networked vehicle and the intelligent networked vehicle immediately ahead after the driving operation is recorded as D safe ; The longitudinal distance between the target intelligent networked vehicle bumper and the bumper of the next intelligent networked vehicle after the target intelligent networked vehicle performs the driving operation is recorded as D back ; The safe distance between the target intelligent networked vehicle and the intelligent networked vehicle immediately behind it after the driving operation is recorded as D back_safe ; Where m represents the number of all intelligent connected vehicles;

[0019] When D safe <D<2D safe , it is considered that the target intelligent connected vehicle is involved in a collision event; at the same time, it is judged that D back and D back_safe The size relationship; when D back >D back_safe When the speed V of the target intelligent connected vehicle at the next moment is updated by the cooperative adaptive cruise control system rec , and identify the feasible driving operations of the target intelligent connected vehicle at the next moment according to the preset driving step length and driving direction; when D back <D back_safe , the next driving operation set at this time will not be considered;

[0020] When D>2D safe When , it is considered that the target intelligent networked vehicle is not involved in a collision event, and the target intelligent networked vehicle maintains the current lane, and the speed V of the target intelligent networked vehicle at the next moment is updated through the cooperative adaptive cruise control system. rec , according to the preset +1 or +2 driving step length, identify the feasible driving operation of the target intelligent connected vehicle at the next moment;

[0021] When D <D safe , the next driving operation set at this time will not be considered;

[0022] Based on the identification process of the feasible driving operation of the target intelligent connected vehicle at the next moment, whether the driving operation of each intelligent connected vehicle at the next moment is feasible is identified respectively, and the feasible driving operation of each intelligent connected vehicle at the next moment is obtained.

[0023] Preferably, the safety distance model includes:

[0024] During the driving process, when the front vehicle suddenly brakes at the maximum deceleration until the speed reaches 0, the rear vehicle can obtain the information of the front vehicle in real time and brake at the maximum deceleration until the speed reaches 0. At this time, the calculation formula for the minimum safe distance that the two vehicles need to maintain when driving is:

[0025] D safe =D F -D L +d0

[0026] Among them, D F Indicates the distance traveled by the rear vehicle during braking. D L Indicates the distance traveled by the front vehicle during braking. d0 represents the minimum safe distance between two vehicles when the maximum deceleration is 0, and d0 is set to 1.5m; D safe Indicates the minimum safe distance that two vehicles need to maintain when driving. This value is a dynamic value related to the speed of the front and rear vehicles and will change with the driving status of the vehicles; a dcc Indicates the maximum deceleration of the vehicle, set a dcc =4.5m / s.

[0027] Preferably, the positions of all the intelligent connected vehicles that are normally traveling on the lanes within the preset range are marked respectively, and recorded as the initial state nodes of the corresponding intelligent connected vehicles; the virtual vehicle positions of the intelligent connected vehicles are obtained according to the feasible driving operations of the intelligent connected vehicles at the next moment, and recorded as the virtual vehicle state nodes of the intelligent connected vehicles; the node sets of the intelligent connected vehicles are obtained according to the initial state nodes of the intelligent connected vehicles and the corresponding virtual vehicle state nodes, and the operation sequence sets of the intelligent connected vehicles are obtained, including:

[0028] Mark the positions of all the intelligent networked vehicles running normally on the preset range of lanes, and record them as the initial state nodes of the corresponding intelligent networked vehicles, that is, (Z1,0), (Z2,0), ..., (Z i ,0),...,(Z m ,0), where m represents the number of all intelligent connected vehicles; Z1, Z2, ..., Z i , ..., Z m represents the number of the intelligent networked vehicle, then the set of all intelligent networked vehicles is H = {Z1, Z2, ..., Z i , ..., Z m};

[0029] According to each feasible driving operation of each intelligent connected vehicle at the next moment, each virtual vehicle position of each intelligent connected vehicle is obtained and recorded as the virtual vehicle state node of each intelligent connected vehicle, that is, the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z1 are (Z1,1), (Z1,2), (Z1,3)...(Z1,a); the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z2 are (Z2,1), (Z2,2), (Z2,3)...(Z2,b); ...; the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z iThe corresponding virtual vehicle state node is (Z i ,1),(Z i ,2),(Z i ,3)...(Z i ,n); ...; Intelligent networked vehicle Z m The corresponding virtual vehicle state node is (Z m ,1),(Z m ,2),(Z m ,3)...(Z m ,M);

[0030] According to the initial state nodes of each intelligent connected vehicle and the corresponding virtual vehicle state nodes, the node set of each intelligent connected vehicle is obtained, that is,

[0031] The node set of the intelligent networked vehicle Z1 is {(Z1,0), (Z1,1), (Z1,2), (Z1,3)...(Z1,a)}; the node set of the intelligent networked vehicle Z2 is {(Z2,0), (Z2,1), (Z2,2), (Z2,3)...(Z2,b)}; ...; the node set of the intelligent networked vehicle Z i The node set is {(Z i ,0),(Z i ,1),(Z i ,2),(Z i ,3)...(Z i ,n)}; ...; Intelligent networked vehicle Z m The node set is {(Z m ,0),(Z m ,1),(Z m ,2),(Z m ,3)...(Z m ,M)}; let the set of all intelligent connected vehicle state nodes be N;

[0032] According to the node set of each intelligent connected vehicle, the operation sequence set of each intelligent connected vehicle is obtained, that is,

[0033] The operation sequence set of the intelligent connected vehicle Z1 is: The operation sequence set of the intelligent connected car Z2 is: Smart Connected Car Z i The set of operation sequences is Smart Connected Car Z m The set of operation sequences is Where a represents the number of feasible driving operations of the intelligent networked vehicle Z1; b represents the number of feasible driving operations of the intelligent networked vehicle Z2; ...; n represents the number of feasible driving operations of the intelligent networked vehicle Z i The number of feasible driving operations; ...; M represents the number of intelligent connected vehicles Zm The number of possible driving operations.

[0034] Preferably, the connection arc is used to represent the connection between each node in each intelligent connected vehicle node set and each node in the remaining intelligent connected vehicle node sets, and the composite network of collision avoidance behaviors of connected vehicles is established according to all intelligent connected vehicle state nodes and connection arcs, including:

[0035] The connection arc is used to represent the connection between each node in each intelligent networked vehicle node set and each node in the rest of the intelligent networked vehicle node sets, then Z i The connection arc between each node in the node set and each node in the rest of the intelligent connected vehicle node set is

[0036] ((Z i ,0),(Z1,0)),((Z i ,0),(Z1,1)),((Z i ,0),(Z1,2)),...,((Z i ,0),(Z1,a));((Z i ,0),(Z2,0)),((Z i ,0),(Z2,1)),((Z i ,0),(Z2,2)),...,((Z i ,0),(Z2,b));...;((Z i ,0),(Z m ,1)),((Z i ,0),(Z m ,1)),((Z i ,0)(Z m ,2)),...,((Z i ,0),(Z m ,M));

[0037] ((Z i ,1),(Z1,0)),((Z i ,1),(Z1,1)),((Z i ,1),(Z1,2)),...,((Z i ,1),(Z1,a));((Z i ,1),(Z2,0)),((Z i ,1),(Z2,1)),((Z i ,1),(Z2,2)),...,((Z i ,1),(Z2,b));...;((Z i ,1),(Z m ,1)),((Z i ,1),(Zm , 1)), ((Z i , 1), (Z m , 2)),..., ((Z i , 1), (Z m , M));

[0038] ((Z i , 2), (Z1, 0)), ((Z i , 2), (Z1, 1)), ((Z i , 2), (Z1, 2)),..., ((Z i , 2), (Z1, a)); ((Z i , 2), (Z2, 0)), ((Z i , 2)(Z2, 1)), ((Z i , 2), (Z2, 2)),..., ((Z i , 2), (Z2, b));...; ((Z i , 2), (Z m , 1)), ((Z i , 2), (Z m , 1)), ((Z i , 2), (Z m , 2)),..., ((Z i , 2), (Z m , M)); ...;

[0040] ((Z i , n), (Z1, 0)), ((Z i , n), (Z1, 1)), ((Z i , n), (Z1, 2)),..., ((Z i , n), (Z1, a)); ((Z i , n), (Z2, 0)), ((Z i , n), (Z2, 1)), ((Z i , n), (Z2, 2)),..., ((Z i , n), (Z2, b));...; ((Z i , n), (Z m , 1)), ((Z i , n), (Z m , 1)), ((Z i , n), (Z m , 2)),..., ((Z i , n), (Z m , M));

[0041] Wherein, i ≠ m;

[0042] According to the above Z i The representation form of the connection arc is used to obtain the connection arc set E of all intelligent connected vehicles;

[0043] According to the state node set N and connection arc E of all intelligent connected vehicles, a composite network G(N,E) of connected vehicle collision avoidance behavior is established.

[0044] Preferably, the attributes defining the connection arc include:

[0045] According to the negative benefits generated by the lane-changing behavior on the upstream and downstream vehicles in the current lane and the target lane, that is, the driving cost of the intelligent connected vehicle, the properties of the connection arc are defined, and the expression is as follows:

[0046]

[0047] Among them, w (v,d),(u,r) (Z i , O, T) represents the impact of the target intelligent networked vehicle’s operation from the current lane to the target lane on the arc ((v, d), (u, r)) on the target intelligent networked vehicle’s benefit; O and T represent the target intelligent networked vehicle Z i For the current lane and the target lane, if the node after the feasible driving operation is in the same lane as the target intelligent connected vehicle node, O and T are the same; Indicates the acceleration of the target intelligent connected vehicle traveling in the current lane; Indicates the acceleration of the target intelligent connected vehicle traveling in the target lane; Represents the acceleration of the intelligent connected vehicle behind the target intelligent connected vehicle after the feasible driving operation is performed; α T Indicates the acceleration of the intelligent connected vehicle behind the target lane before executing the feasible driving maneuver; Indicates the acceleration of the intelligent connected vehicle behind the current lane after the feasible driving operation is performed; α O The acceleration of the connected vehicle behind the current lane before the feasible driving maneuver is performed;

[0048] It indicates that the speed advantage obtained by the target intelligent connected vehicle through lane changing has a positive impact on the target intelligent connected vehicle’s revenue; η is the politeness factor; It indicates the impact of executing feasible driving operations on subsequent intelligent connected vehicles to avoid damaging traffic stability; μ is the benefit factor; Indicates the speed advantage of subsequent intelligent connected vehicles in the current lane due to the target intelligent connected vehicle performing feasible driving operations;

[0049] Through the coordinated adaptive cruise control system, the values ​​of the above accelerations can be obtained, thereby obtaining Z iThe values ​​of all connection arcs; and then the values ​​of all connection arcs in the composite network are obtained.

[0050] Preferably, in the composite network, a decision function is set, and the target function of the target intelligent networked vehicle is established by combining the connection arcs; constraint conditions are set; and the minimum value of the target function is calculated according to a preset number of iterations. At this time, the driving route composed of all the connection arcs and nodes obtained is the optimal driving route of the target intelligent networked vehicle, including:

[0051] In the composite network G(N,E), if the target intelligent connected vehicle is at the node (v,d), there will be a travel cost between the node (v,d) and the next node (u,r) connected to it, that is, the connection arc indicates that there is a negative benefit between the two nodes (v,d) and (u,r);

[0052] Set the decision function x (v,d),(u,r) , combined with the connection arc, the objective function of deciding the optimal driving route is established, and its expression is

[0053]

[0054] Among them, x (v,d),(u,r) ∈{0,1}, when x (v,d),(u,r) When it is 1, it means the arc ((v,d),(u,r)) is selected; when x (v,d),(u,r) When it is 0, it means that the arc ((v,d), (u,r)) is not selected; E represents the set of arcs;

[0055] When the connection arc between the front and back nodes is selected, the front and back nodes must be selected. Based on this premise, constraints are set at both ends of the arc to ensure the balance of traffic flow in the composite network. Constraints are set, including:

[0056] Assuming that the current position of the target intelligent connected vehicle is (v, d), the next position is (u, r), and the constraint condition of the decision variable of node (v, d) is set as

[0057]

[0058] Among them, y (v,d) ∈{0,1}, when y (v,d) When 1 is selected, it indicates that the node (v, d) is selected; when y (v,d) When it is 0, it means that the node (v, d) is not selected; N represents the node set; this formula means that when selecting the next node on the (v, d) node, for any selectable node (u, r), y (v,d) Must not be greater than all x (v,d),(u,r) sum;

[0059] Assuming that the current position of the target intelligent connected vehicle is (u, r), the previous position is (v, d), and the constraint condition of the decision variable of node (u, r) is set as

[0060]

[0061] Among them, y (u,r) ∈{0,1}, when y (u,r) When 1 is selected, it indicates that the node (u, r) is selected; when y (u,r) When it is 0, it means that the node (u, r) is not selected; N represents the node set; this formula means that on the node (u, r), when the previous node is selected, for any selectable node (v, d), y (u,r) Must not be greater than all x (v,d),(u,r) sum;

[0062] In the case of multi-vehicle cooperative driving, in order to avoid vehicle conflicts, the sum of the decision functions of each step corresponding to the target vehicle intelligent connected vehicle is stipulated to be 1, then the constraint condition is expressed as

[0063]

[0064] Among them, Z i represents the target intelligent connected vehicle number; H represents the set of all intelligent connected vehicles; f represents the number of feasible driving operations; The target intelligent connected vehicle Z i A set of operation sequences;

[0065] According to all the connection arcs and nodes in the composite network, different driving routes of the target intelligent connected vehicle are obtained when iterating until a preset number of iterations is reached. According to the value of each connection arc in each driving route, the value of the sum of all the connection arcs in each driving route is calculated. By comparing the values ​​of the sum of all the connection arcs, the objective function is solved to obtain the driving route corresponding to the minimum sum of all the connection arcs of the target intelligent connected vehicle when iterating until the preset number of iterations. At this time, the driving route composed of all the connection arcs and nodes is the optimal driving route of the target intelligent connected vehicle.

[0066] Preferably, the number of iterations is set to 200.

[0067] The present invention also provides a multi-vehicle cooperative collision avoidance device for intelligent connected vehicles in a mixed flow environment based on a composite network, comprising:

[0068] Memory for storing computer programs;

[0069] A processor is used to implement the steps of the method for multi-vehicle collaborative collision avoidance of intelligent connected vehicles in a composite network mixed flow environment as described above when executing the computer program.

[0070] The present invention also provides a computer-readable storage medium, including: a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method for collaborative collision avoidance of multiple intelligent connected vehicles in a composite network mixed flow environment are implemented.

[0071] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0072] (1) The method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network described in the present invention uses a mathematical modeling method to break away from the framework of logical judgment. It does not require redesigning the model for different traffic scenarios, is applicable to most traffic scenarios, has a wider range of applications, and is easy to transplant.

[0073] (2) The present invention discloses a method for multi-vehicle cooperative collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network. The method takes intelligent connected vehicles as the research object, considers the impact on vehicles in the front and rear range of a blocked road section, and transforms the collision avoidance problem of connected vehicles into a new strategy for network flow optimization problem. When an accident occurs in front of the road, it helps the traffic behind to change lanes in advance and select the optimal planned route; the traffic within the accident range cooperates to avoid collision and quickly passes through the blocked road section; the traffic behind the accident quickly leaves to provide sufficient space for vehicles that have just passed through the blocked road section to avoid the expansion of the blocked range; the vehicles cooperate to operate and make full use of CACC technology to avoid traffic congestion;

[0074] (3) The method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network described in the present invention constructs a virtual composite network based on real traffic flow, optimizes the composite network, makes the traffic flow faster, more efficient, and more closely connected, thereby improving road traffic efficiency and traffic environment;

[0075] (4) The method for collaborative collision avoidance of multiple intelligent connected vehicles in a composite network mixed flow environment described in the present invention integrates network model construction, optimal decision model and simulation verification into one, fully verifies the effectiveness and feasibility of the model, and makes the technical framework more complete. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0077] Figure 1 It is a flow chart of a multi-vehicle collaborative collision avoidance method for intelligent connected vehicles in a composite network mixed flow environment provided by the present invention;

[0078] Figure 2 It is a schematic diagram of the safety distance calculation process provided by the present invention;

[0079] Figure 3 It is a schematic diagram between the target intelligent network-connected vehicle and other intelligent network-connected vehicles provided by the present invention;

[0080] Figure 4 It is a schematic diagram of a set of operation sequences of the intelligent networked vehicle provided by the present invention;

[0081] Figure 5 It is a schematic diagram of constructing a target intelligent networked vehicle node and network provided by the present invention;

[0082] Figure 6 The present invention provides a composite network G(N,E) for building a networked vehicle. B )

[0083] Figure 7 The present invention provides a structure G (N, E B )

[0084] Figure 8 It is a comparison chart of model simulation data provided by the present invention;

[0085] Fig. 9 It is a time point scatter plot of the minimum measured value of the collision time provided by the present invention changing with time;

[0086] Description of the Figures in the Specification: Figure 3 (a) shows a vehicle breakdown on the road, and the position of the vehicle is shown in the figure; Figure 3 (b) is a schematic diagram showing the target intelligent connected vehicle establishing connections with surrounding CAVs. DETAILED DESCRIPTION

[0087] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0088] Reference Figure 1 As shown, Figure 1 This is a schematic diagram of a multi-vehicle collaborative collision avoidance method for intelligent connected vehicles in a composite network mixed flow environment provided by the present invention; the specific steps include:

[0089] Step 1: When a vehicle breaks down on the road, the smart connected vehicles on the road sections before and after the accident need to choose a route with a faster speed, closer connection with other smart connected vehicles around, and a smaller speed difference, so as to encourage the smart connected vehicles to apply the CACC mode to the greatest extent and quickly pass through the congested road section;

[0090] When a faulty vehicle appears on a lane during the driving of an intelligent connected vehicle and a human-driven vehicle, the next driving operation of each intelligent connected vehicle is set for all the intelligent connected vehicles that are normally driving on the lanes within a preset range; based on the collaborative adaptive cruise control system and the safety distance model, it is identified whether the next driving operation of each intelligent connected vehicle is feasible, and the next feasible driving operation of each intelligent connected vehicle is obtained;

[0091] That is, according to the preset driving step length and driving direction, the next moment driving operation of each intelligent networked vehicle is set for all the intelligent networked vehicles that are normally driving on the lanes within the preset range;

[0092] The preset driving step lengths are +1, -1, +2, etc.; the preset driving directions are forward, backward, left, right, etc.; the preset range of lanes can be three lanes, five lanes, etc., and the lane length can be 2000 meters; the set next moment driving operation includes: straight driving operation and lane change driving operation;

[0093] Based on the cooperative adaptive cruise control system, for the target intelligent connected vehicle Z i , i = 1, 2, 3..., m, the longitudinal distance between the target intelligent networked vehicle bumper and the bumper of the intelligent networked vehicle immediately ahead after the driving operation is recorded as D; the safety distance between the target intelligent networked vehicle and the intelligent networked vehicle immediately ahead after the driving operation is recorded as D safe ; The longitudinal distance between the target intelligent networked vehicle bumper and the bumper of the intelligent networked vehicle immediately behind it after the driving operation is recorded as D back ; The safe distance between the target intelligent networked vehicle and the intelligent networked vehicle immediately behind it after the driving operation is recorded as D back_safe ; Where m represents the number of all intelligent connected vehicles;

[0094] When D safe <D<2D safe , it is considered that the target intelligent connected vehicle is involved in a collision event; at the same time, it is judged that D back and D back_safe The size relationship; when D back >D back_safe When the speed V of the target intelligent connected vehicle at the next moment is updated by the cooperative adaptive cruise control system rec , and identify the feasible driving operations of the target intelligent connected vehicle at the next moment according to the preset driving step length and driving direction; when D back <D back_safe , the next driving operation set at this time will not be considered;

[0095] When D>2D safeWhen , it is considered that the target intelligent networked vehicle is not involved in a collision event, and the target intelligent networked vehicle maintains the current lane, and the speed V of the target intelligent networked vehicle at the next moment is updated through the cooperative adaptive cruise control system. rec , according to the preset +1 or +2 driving step length, identify the feasible driving operation of the target intelligent connected vehicle at the next moment;

[0096] When D <D safe , the next driving operation set at this time will not be considered;

[0097] Based on the identification process of the feasible driving operation of the target intelligent connected vehicle at the next moment, whether the driving operation of each intelligent connected vehicle at the next moment is feasible is identified respectively, and the feasible driving operation of each intelligent connected vehicle at the next moment is obtained;

[0098] Among them, the safety distance model includes:

[0099] During the driving process, when the front vehicle suddenly brakes at the maximum deceleration until the speed reaches 0, the rear vehicle can obtain the information of the front vehicle in real time and brake at the maximum deceleration until the speed reaches 0. At this time, the calculation formula for the minimum safe distance that the two vehicles need to maintain when driving is:

[0100] D safe =D F -D L +d0

[0101] Among them, D F Indicates the distance traveled by the rear vehicle during braking. D L Indicates the distance traveled by the front vehicle during braking. d0 represents the minimum safe distance between two vehicles when the maximum deceleration is 0, and d0 is set to 1.5m; D safe Indicates the minimum safe distance that two vehicles need to maintain when driving. This value is a dynamic value related to the speed of the front and rear vehicles and will change with the driving status of the vehicles; a dcc Indicates the maximum deceleration of the vehicle, set a dcc =4.5m / s.

[0102] Step 2: Mark the positions of all the intelligent connected vehicles running normally on the preset range of lanes, and record them as the initial state nodes of the corresponding intelligent connected vehicles, namely (Z1,0), (Z2,0), ..., (Z i ,0),...,(Z m ,0), where m represents the number of all intelligent connected vehicles; Z1, Z2, ..., Z i , ..., Z m represents the number of the intelligent networked vehicle, then the set of all intelligent networked vehicles is H = {Z1, Z2, ..., Z i, ..., Z m};

[0103] According to each feasible driving operation of each intelligent connected vehicle at the next moment, each virtual vehicle position of each intelligent connected vehicle is obtained and recorded as the virtual vehicle state node of each intelligent connected vehicle, that is, the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z1 are (Z1,1), (Z1,2), (Z1,3)...(Z1,a); the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z2 are (Z2,1), (Z2,2), (Z2,3)...(Z2,b); ...; the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z i The corresponding virtual vehicle state node is (Z i ,1),(Z i ,2),(Z i ,3)...(Z i ,n); ...; Intelligent networked vehicle Z m The corresponding virtual vehicle state node is (Z m ,1),(Z m ,2),(Z m ,3)...(Z m ,M);

[0104] According to the initial state nodes of each intelligent connected vehicle and the corresponding virtual vehicle state nodes, the node set of each intelligent connected vehicle is obtained, that is,

[0105] The node set of the intelligent networked vehicle Z1 is {(Z1,0), (Z1,1), (Z1,2), (Z1,3)...(Z1,a)}; the node set of the intelligent networked vehicle Z2 is {(Z2,0), (Z2,1), (Z2,2), (Z2,3)...(Z2,b)}; ...; the node set of the intelligent networked vehicle Z i The node set is {(Z i ,0),(Z i ,1),(Z i ,2),(Z i ,3)...(Z i ,n)}; ...; Intelligent networked vehicle Z m The node set is {(Z m ,0),(Z m ,1),(Z m ,2),(Z m ,3)...(Z m ,M)}; let the set of all intelligent connected vehicle state nodes be N;

[0106] According to the node set of each intelligent connected vehicle, the operation sequence set of each intelligent connected vehicle is obtained, that is,

[0107] The operation sequence set of the intelligent connected vehicle Z1 is: The operation sequence set of the intelligent connected car Z2 is: Smart Connected Car Z i The set of operation sequences is Smart Connected Car Z m The set of operation sequences is Where a represents the number of feasible driving operations of the intelligent networked vehicle Z1; b represents the number of feasible driving operations of the intelligent networked vehicle Z2; ...; n represents the number of feasible driving operations of the intelligent networked vehicle Z i The number of feasible driving operations; ...; M represents the number of intelligent connected vehicles Z m the number of feasible driving operations;

[0108] The connection arc is used to represent the connection between each node in each intelligent networked vehicle node set and each node in the rest of the intelligent networked vehicle node sets, then Z i The connection arc between each node in the node set and each node in the rest of the intelligent connected vehicle node set is

[0109] ((Z i ,0),(Z1,0)),((Z i ,0),(Z1,1)),((Z i ,0),(Z1,2)),...,((Z i ,0),(Z1,a));((Z i ,0),(Z2,0)),((Z i ,0),(Z2,1)),((Z i ,0),(Z2,2)),...,((Z i ,0),(Z2,b));...;((Z i ,0),(Z m ,1)),((Z i ,0),(Z m ,1)),((Z i ,0)(Z m ,2)),...,((Z i ,0),(Z m ,M));

[0110] ((Z i ,1),(Z1,0)),((Z i ,1),(Z1,1)),((Z i ,1),(Z1,2)),...,((Z i ,1),(Z1,a));((Z i ,1),(Z2,0)),((Z i,1),(Z2,1)),((Z i ,1),(Z2,2)),...,((Z i ,1),(Z2,b));...;((Z i ,1),(From m ,1)),((From i ,1),(From m ,1)),((From i ,1),(From m ,2)),...,((From i ,1),(From m ,M));

[0111] ((WITH i ,2),(Z1,0)),((Z i ,2),(Z1,1)),((Z i ,2),(Z1,2)),...,((Z i ,2),(Z1,a));((Z i ,2),(Z2,0)),((Z i ,2)(Z2,1)),((Z i ,2),(Z2,2)),...,((Z i ,2),(Z2,b));...;((Z i ,2),(From m ,1)),((From i ,2),(From m ,1)),((From i ,2),(From m ,2)),...,((From i ,2),(From m ,M)); ...;

[0113] ((WITH i ,n),(Z1,0)),((Z i ,n),(Z1,1)),((Z i ,n),(Z1,2)),...,((Z i ,n),(Z1,a));((Z i ,n),(Z2,0)),((Z i ,n),(Z2,1)),((Z i ,n),(Z2,2)),...,((Z i ,n),(Z2,b));...;((Z i ,n),(Z m ,1)),((From i,n),(Z m ,1)),((Z i ,n),(Z m ,2)),...,((Z i ,n),(Z m ,M));

[0114] Among them, i≠m;

[0115] According to the above Z i The representation form of the connection arc is used to obtain the connection arc set E of all intelligent connected vehicles;

[0116] According to the state node set N and the connection arc E of all intelligent connected vehicles, a composite network G(N,E) of connected vehicle collision avoidance behavior is established;

[0117] According to step 1 and step 2, the feasible driving operations obtained are in parallel with the initial state of the intelligent connected vehicle. Therefore, when considering the next node of the current intelligent connected vehicle, it is necessary to select from different node sets; each intelligent connected vehicle can locate the position of the intelligent connected vehicle through sensors and information interaction between them.

[0118] Step 3: Based on the negative benefits generated by the lane-changing behavior on the upstream and downstream vehicles in the current lane and the target lane, that is, the driving cost of the intelligent connected vehicle, define the attributes of the connection arc, and its expression is as follows:

[0119]

[0120] Among them, w (v,d),(u,r) (Z i , O, T) represents the impact of the target intelligent networked vehicle’s operation from the current lane to the target lane on the arc ((v, d), (u, r)) on the target intelligent networked vehicle’s benefit; O and T represent the target intelligent networked vehicle Z i For the current lane and the target lane, if the node after the feasible driving operation is in the same lane as the target intelligent connected vehicle node, O and T are the same; Indicates the acceleration of the target intelligent connected vehicle traveling in the current lane; Indicates the acceleration of the target intelligent connected vehicle traveling in the target lane; Represents the acceleration of the intelligent connected vehicle behind the target intelligent connected vehicle after the feasible driving operation is performed; α T Indicates the acceleration of the intelligent connected vehicle behind the target lane before executing the feasible driving maneuver; Indicates the acceleration of the intelligent connected vehicle behind the current lane after the feasible driving operation is performed; α O The acceleration of the connected vehicle behind the current lane before the feasible driving maneuver is performed;

[0121] It indicates that the speed advantage obtained by the target intelligent connected vehicle through lane changing has a positive impact on the target intelligent connected vehicle’s revenue; η is the politeness factor; It indicates the impact of executing feasible driving operations on subsequent intelligent connected vehicles to avoid damaging traffic stability; μ is the benefit factor; Indicates the speed advantage of subsequent intelligent connected vehicles in the current lane due to the target intelligent connected vehicle performing feasible driving operations;

[0122] Through the coordinated adaptive cruise control system, the values ​​of the above accelerations can be obtained, thereby obtaining Z i The values ​​of all connection arcs; and then the values ​​of all connection arcs in the composite network are obtained.

[0123] Step 4: During driving, the intelligent connected vehicle can receive driving information from surrounding intelligent connected vehicles in real time through on-board sensors. By analyzing the collected driving information, the possible negative impact of surrounding intelligent connected vehicles on the current intelligent connected vehicle can be evaluated, and timely response measures can be taken to improve traffic efficiency while ensuring vehicle driving safety. In order to improve traffic efficiency and effectively alleviate traffic congestion when an accident occurs ahead, the intelligent connected vehicle needs to find the optimal driving route in the composite network.

[0124] That is, in the composite network G(N,E), if the target intelligent connected vehicle is at the node (v,d), there will be a travel cost between the node (v,d) and the next node (u,r) connected to it, that is, the connection arc indicates that there is a negative benefit between the two nodes (v,d) and (u,r);

[0125] Set the decision function x (v,d),(u,r) , combined with the connection arc, the objective function of deciding the optimal driving route is established, and its expression is

[0126]

[0127] Among them, x (b,d),(u,r) ∈{0,1}, when x (v,d),(u,r) When it is 1, it means the arc ((v,d),(u,r)) is selected; when x (v,d),(u,r) When it is 0, it means that the arc ((v,d), (u,r)) is not selected; E represents the set of arcs;

[0128] When the connection arc between the front and back nodes is selected, the front and back nodes must be selected. Based on this premise, constraints are set at both ends of the arc to ensure the balance of traffic flow in the composite network. Constraints are set, including:

[0129] Assuming that the current position of the target intelligent connected vehicle is (v, d), the next position is (u, r), and the constraint condition of the decision variable of node (v, d) is set as

[0130]

[0131] Among them, y (v,d) ∈{0,1}, when y (v,d) When 1 is selected, it indicates that the node (v, d) is selected; when y (v,d) When it is 0, it means that the node (v, d) is not selected; N represents the node set; this formula means that when selecting the next node on the (v, d) node, for any selectable node (u, r), y (v,d) Must not be greater than all x (v,d),(u,r) sum;

[0132] Assuming that the current position of the target intelligent connected vehicle is (u, r), the previous position is (v, d), and the constraint condition of the decision variable of node (u, r) is set as

[0133]

[0134] Among them, y (u,r) ∈{0,1}, when y (u,r) When 1 is selected, it indicates that the node (u, r) is selected; when y (u,r) When it is 0, it means that the node (u, r) is not selected; N represents the node set; this formula means that on the node (u, r), when the previous node is selected, for any selectable node (v, d), y (u,r) Must not be greater than all x (v,d),(u,r) sum;

[0135] In the case of multi-vehicle cooperative driving, in order to avoid vehicle conflicts, the sum of the decision functions of each step corresponding to the target vehicle intelligent connected vehicle is stipulated to be 1, then the constraint condition is expressed as

[0136]

[0137] Among them, Z i represents the target intelligent connected vehicle number; H represents the set of all intelligent connected vehicles; f represents the number of feasible driving operations; The target intelligent connected vehicle Z i A set of operation sequences;

[0138] According to all the connection arcs and nodes in the composite network, different driving routes of the target intelligent networked vehicle when iterating until a preset number of iterations is reached are obtained, and according to the value of each connection arc in each driving route, the value of the sum of all the connection arcs in each driving route is calculated, and by comparing the values ​​of the sum of all the connection arcs, the objective function is solved to obtain the driving route corresponding to the minimum sum of all the connection arcs of the target intelligent networked vehicle when iterating until the preset number of iterations. At this time, the driving route composed of all the connection arcs and nodes is the optimal driving route of the target intelligent networked vehicle;

[0139] The meaning of the objective function is "to select the operation with the least negative impact among the currently feasible operations and which can make the most of the use of the underlying CACC model". Through continuous iteration, the optimal driving route can be selected. In the case of a traffic accident assuming that there is a traffic accident ahead, the optimal driving route is to actively change lanes and avoid collisions, and coordinate with other CAVs to quickly pass through congested sections, thereby improving traffic efficiency.

[0140] Step 5: Based on the acquisition process of the optimal driving route of the above-mentioned target intelligent connected vehicle, the optimal driving routes of all intelligent connected vehicles are obtained in turn, so that all intelligent connected vehicles that are driving normally on the road behind the faulty intelligent connected vehicle can change lanes in advance according to their corresponding optimal driving routes to avoid the preset influence range of the faulty intelligent connected vehicle; the intelligent connected vehicles adjacent to the faulty intelligent connected vehicle can quickly pass through the preset influence range of the faulty intelligent connected vehicle; all intelligent connected vehicles that are driving normally on the road in front of the faulty intelligent connected vehicle can quickly pass through in coordination.

[0141] In a specific embodiment of the present invention, according to the vehicle networking technology, a virtual connection network is constructed between the intelligent networked vehicles traveling on the road, and the connection arcs are given corresponding attributes to select the optimal route (the minimum cost flow problem in operations research). In this way, the multi-vehicle collision avoidance problem is transformed into a network flow optimization problem, and the traffic congestion problem is alleviated by making the network more efficient, safe, fast, compact, and better connected. Taking the intelligent networked vehicle B as the target intelligent networked vehicle, the specific steps of constructing a partial composite network related to vehicle B are as follows:

[0142] Step 1: Under the new mixed flow condition where CAVs and human-driven vehicles (HVs) share a lane, when a vehicle in front breaks down, the CAVs driving normally in the lane will complete the lane change to avoid collision. Among them, the CAVs using CACC exchange information and coordinate lane change, which greatly improves the lane traffic efficiency and reduces the possibility of traffic congestion caused by the accident ahead.

[0143] like Figure 3 As shown in (a), Figure 3 The red vehicle in (a) indicates a traffic accident in front of the target three lanes, the red shadow indicates the impact area of ​​the traffic accident, and the green and purple vehicles represent CAVs and HVs driving normally on the road behind, respectively; CAVs and HVs will perform collision avoidance actions to bypass the impact area of ​​the accident and avoid traffic congestion;

[0144] Mark the positions of CAVs A, B, C, D, and E on the three lanes behind, which are equipped with CACC devices, as (A, 0), (B, 0), (C, 0), (D, 0), and (E, 0); build an information interaction network; take CAV B as the target CAV, and the connection between CAV B and the surrounding CAVs is represented by arcs ((B, 0), (A, 0)), ((B, 0), (C, 0)), ((B, 0), (D, 0)), and ((B, 0), (E, 0)). Figure 3 (b) as shown.

[0145] Step 2: Before the CAVs make the optimal driving strategy, there are multiple feasible driving operations at the next moment starting from the current position; finding the optimal lane change strategy through the feasible driving operations is the basis for deciding the optimal route; in order to facilitate observation and modeling, in this embodiment, two feasible driving operations at the next moment are set for each CAV on the lane;

[0146] For any intelligent connected vehicle, the set of different state nodes of the same vehicle is the set of operation sequences of the vehicle; Figure 4 As shown in the figure, the green vehicle represents the target intelligent connected vehicle. When the target intelligent connected vehicle is driving on three lanes, there are a left front vehicle (LF), a left rear vehicle (LB), a right front vehicle (RF) and a right rear vehicle (RB) around the target intelligent connected vehicle; operation 0 represents the current position operation of the target intelligent connected vehicle, operation 1 represents the forward driving operation of the intelligent connected vehicle, operation 2 represents the driving operation of the target intelligent connected vehicle to change lanes to the left, and operation 3 represents the driving operation of the intelligent connected vehicle to change lanes to the right. Then, operation 0, operation 1, operation 2 and operation 3 are the operation set {0, 1, 2, 3} of the intelligent connected vehicle;

[0147] like Figure 5 As shown, the light green shadow (B,1) and the dark green shadow (B,2) respectively represent the corresponding virtual vehicles obtained after the intelligent connected vehicle B executes the feasible driving operation at the next moment, then (B,0) represents the initial state, (B,1) represents the driving operation 1 of vehicle B, and (B,2) represents the driving operation 2 of vehicle B; (A,0), (C,0), (D,0), (E,0) represent the initial states of vehicle A, vehicle C, vehicle D, and vehicle E in turn; (A,1) represents the driving operation 1 of vehicle A, and (A,2) represents the driving operation 2 of vehicle A; (C,1) represents the driving operation 1 of vehicle C, and (C,2) represents the driving operation 2 of vehicle C; (D,1) represents the driving operation 1 of vehicle D, and (D,2) represents the driving operation 2 of vehicle D; (E,1) represents the driving operation 1 of vehicle E, and (E,2) represents the driving operation 2 of vehicle E;

[0148] According to the feasible driving operations of each intelligent connected vehicle at the next moment, the virtual vehicle positions of each intelligent connected vehicle are obtained and recorded as the virtual vehicle state nodes of each intelligent connected vehicle; the initial state nodes and virtual vehicle state nodes of the intelligent connected vehicles are regarded as nodes in the network, and the node sets corresponding to each intelligent connected vehicle are obtained. The node set of vehicle A is {(A,0), (A,1), (A,2)}, the node set of vehicle B is {(B,0), (B,1), (B,2)}, the node set of vehicle C is {(C,0), (C,1), (C,2)}, the node set of vehicle D is {(D,0), (D,1), (D,2)}, and the node set of vehicle E is {(E,0), (E,1), (E,2)}; the node set of all intelligent connected vehicles is recorded as N;

[0149] According to the description of the above intelligent connected vehicle operation set, in this embodiment, the operation sequence set of vehicle A is {0,1,2}, the operation sequence set of vehicle B is {0,1,2}, the operation sequence set of vehicle C is {0,1,2}, the operation sequence set of vehicle D is {0,1,2}, and the operation sequence set of vehicle E is {0,1,2}; Figure 5 As shown, there are 5 sets of operation sequences in total; Figure 5 The solid line in the middle represents the connection arc between car B and A, C, D, and E; the dotted line represents the connection arc between the initial state node (B, 0) of car B and the virtual vehicle nodes corresponding to A, C, D, and E, that is, Figure 5 The connecting lines between (B,0) and (A,1), (A,2), (C,1), (C,2), (D,1), (D,2), (E,1), (E,2);

[0150] In order to achieve the goal of multi-vehicle collaborative collision avoidance of CAVs in the network, each CAV should consider the connection between the virtual vehicle state node obtained by driving operation and the other intelligent connected vehicles and their corresponding virtual vehicles before making a lane change decision; in this embodiment, taking the target intelligent connected vehicle B as an example, while considering the connection between the current state node (B,0) of the intelligent connected vehicle B and other CAVs, it is also necessary to build the connection between the operation nodes (B,1), (B,2) and other CAVs and their corresponding virtual vehicles;

[0151] Reference Figure 6 As shown, Figure 6The dashed lines connecting (B,1) and (A,0), (A,1), (A,2), (C,0), (C,1), (C,2), (D,0), (D,1), (D,2), (E,0), (E,1), (E,2) indicate the connection arcs between (B,1) and (A,0), (A,1), (A,2), (C,0), (C,1), (C,2), (D,0), (D,1), (D,2), (E,0), (E,1), (E,2) ; The dotted line connecting (B,2) and (A,0), (A,1), (A,2), (C,0), (C,1), (C,2), (D,0), (D,1), (D,2), (E,0), (E,1), (E,2) represents the connection arc between (B,2) and (A,0), (A,1), (A,2), (C,0), (C,1), (C,2), (D,0), (D,1), (D,2), (E,0), (E,1), (E,2);

[0152] Reference Figure 7 As shown in Figure 2, the obtained composite network can be decomposed into three parts, which respectively represent the connections between the three state nodes of vehicle B and the state nodes of other intelligent connected vehicles. The set of all connection arcs of vehicle B is denoted as E. B ; Figure 7 All nodes and arcs in the network constitute the composite network G(N,E) related to vehicle B. B ).

[0153] In a specific embodiment of the present invention, the model is numerically simulated and verified using the SUMO simulation platform. The SUMO simulation platform can simulate many actual traffic scenarios. When SUMO is used as a test platform for intelligent traffic control algorithms, it is required to interact well with external programs / algorithms. The Traci interface is used to communicate with the SUMO simulation platform. Communicating with SUMO through languages ​​such as Python can avoid always operating in the graphical interface of sumo-gui, thereby improving work efficiency.

[0154] Assume that the maximum acceleration of all vehicles is a acc =3.5m / s, the maximum deceleration of all vehicles a dcc =4.5m / s; CAV and HV each account for 50% of the vehicles in the traffic flow. The CACC model is used to simulate the driving behavior of CAV, the Wiedemann99 model is used to simulate the behavior of HV, and the collision avoidance model proposed by the present invention is used to optimize the CAV collision avoidance operation when the HV makes an emergency brake; the simulated road is a 2000-meter straight three-lane highway with a road speed limit of 60km / h;

[0155] The present invention uses a CACC system to control the speed and position of a CAV. In this system, a connected-automated vehicle (CAV) shares its own parameters with other CAVs in the network through V2V communication. This is achieved in an autonomous manner without the need for central management. CACC uses the advantages of V2V communication to enable CAVs to form a queue and travel at a coordinated speed with a shorter time interval between each other. By sharing vehicle information such as acceleration, speed and position in a distributed manner, CAVs within a certain communication range can cooperate with other vehicles, thereby achieving advantages such as improved driving safety, increased road capacity, and reduced unnecessary speed changes.

[0156] In the SUMO simulation software, the CACC model is divided into four modes: speed control mode, gap control mode, gap-closing control mode, and collision avoidance control mode, which are applied to the following scenarios respectively:

[0157] Speed ​​control mode: When the headway distance is greater than 2s or there is no vehicle ahead within the vehicle's detection range, the vehicle maintains the driver's desired speed.

[0158] Distance control mode: When the distance between the front and rear vehicles is less than 0.2m or the speed difference is less than 0.1m / s, the vehicle maintains a fixed headway with the vehicle in front.

[0159] Distance proximity control mode: When the headway between the front and rear vehicles is less than 1.5s, this mode is activated to ensure a smooth transition between the speed control mode and the distance control mode.

[0160] Anti-collision mode: This mode is activated when the headway distance is less than 1.5s or the distance deviation is negative.

[0161] The present invention uses the Wiedemann99 model to describe the following behavior of human-driven vehicles (HVs); the Wiedemann99 model is applicable to driving behaviors on urban expressways, suburban roads or highways; in this model, when a high-speed HV approaches a slower vehicle in front, it will start to slow down once it reaches its own safety perception threshold. Due to the imperfect estimation of the speed of the vehicle in front by the driver, the speed may be lower than that of the vehicle in front; due to different physical properties, each HV has different perception thresholds for safety distance, expected speed and speed difference; among them, when tracking using the Wiedemann99 model, at a given speed v, the tracking distance D G=d0+h×v, requiring the expected static distance between the two HVs d0=1.5m and the headway h=0.9s; in addition, the longitudinal oscillation of the HV is expected to be 4 meters, and the speed difference in the "following" state is between -0.35 and 0.35.

[0162] In order to simulate the collision avoidance risk verification and comparison experiment, 9 HVs were placed in the interval of 400 to 800 meters from the starting point to simulate the accident behavior of HVs, and the start time and end time of the accident were set; the accident vehicle decelerated at the maximum deceleration at the start time of the accident, stopped on the road after the speed was reduced to 0, and accelerated at the maximum acceleration at the end time of the accident until the speed reached the safe speed; the accident vehicle numbers and the start and end times of the accident in the experiment are shown in Table 1:

[0163] Table 1 Accident vehicle number and accident start and end time

[0164]

[0165] The faster the vehicle passes through the fault range, the better the performance of the model; the shorter the time the vehicle takes to pass through the current fault range, that is, the less time lost, the better the performance of the model; within the same time, the more vehicles pass through the fault range, the better the performance of the model; therefore, the present invention uses average speed, time loss and flow to compare the advantages and disadvantages of models.

[0166] Detectors are set up on three roads to measure the number of vehicles entering the lane, the number of vehicles leaving the lane, the average speed and the running time loss. The effectiveness of the collision avoidance model is verified by comparing the data obtained by the proposed new collision avoidance model and the traditional CACC model.

[0167] Table 2 Simulation data under the collision avoidance model of the present invention

[0168]

[0169] Table 3 Simulation data under CACC model

[0170]

[0171] After SUMO simulation, the number of vehicles entering the lane, the number of vehicles leaving the lane, the average speed and time loss of the collision avoidance model of the present invention and the traditional CACC model within 3600 seconds of simulation were obtained; Table 2 is the simulation data under the collision avoidance model of the present invention, and Table 3 is the simulation data under the CACC model.

[0172] It can be found from Table 3 that when the collision avoidance model proposed in this paper is applied, when an accident occurs in front, the vehicles driving normally in the lane can change lanes in time to avoid collision and reduce time loss; Figure 8As shown in the figure, after using the collision avoidance model proposed in this paper, the average speed of vehicles is greatly improved, the time loss is reduced, the hourly traffic flow is increased, the traffic pressure is effectively alleviated, and the road capacity is improved.

[0173] In order to study the safety issues in the process of vehicle driving, a scatter plot of the time point of the minimum measured value of the collision time TTC (Time-To-Collision) versus time is drawn according to the simulation experiment results; Fig. 9 As shown in the figure, the time when vehicles in the CACC model collide is close to the shortest collision time point, leaving less time for CAVs to observe and calculate and for human drivers to react and avoid collisions in time, resulting in many densely packed traffic conflict points. However, after using the new collision avoidance model, the minimum TTC measurement value of the vehicle is a certain distance away from the time point when the conflict occurs, which effectively avoids the occurrence of secondary accidents to a certain extent. The traffic conflict points in the figure are scattered and few in number.

[0174] A specific embodiment of the present invention also provides a device for collaborative collision avoidance of multiple intelligent connected vehicles in a composite network mixed flow environment, including: a memory for storing a computer program; and a processor for implementing the steps of the above-mentioned method for collaborative collision avoidance of multiple intelligent connected vehicles in a composite network mixed flow environment when executing the computer program.

[0175] A specific embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for multi-vehicle collaborative collision avoidance of intelligent connected vehicles in a composite network mixed flow environment are implemented.

[0176] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0177] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0178] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0179] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0180] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A multi-vehicle collaborative collision avoidance method for intelligent connected vehicles in a mixed flow environment based on a composite network, characterized in that: include: When a faulty vehicle appears on a lane during the driving of an intelligent networked vehicle and a human-driven vehicle, the next driving operation of each intelligent networked vehicle is set for all the intelligent networked vehicles that are normally driving on the lanes within a preset range; Based on the cooperative adaptive cruise control system, according to the safety distance model, it is identified whether the driving operation of each intelligent connected vehicle at the next moment is feasible, and the feasible driving operation of each intelligent connected vehicle at the next moment is obtained; The positions of all the intelligent connected vehicles that are normally traveling on the lanes within the preset range are marked respectively, and recorded as the initial state nodes of the corresponding intelligent connected vehicles; according to the feasible driving operations of each intelligent connected vehicle at the next moment, the virtual vehicle positions of each intelligent connected vehicle are obtained, and marked as the virtual vehicle state nodes of each intelligent connected vehicle; according to the initial state nodes of each intelligent connected vehicle and the corresponding virtual vehicle state nodes, the node sets of each intelligent connected vehicle are obtained, and the operation sequence sets of each intelligent connected vehicle are obtained; Use connection arcs to represent the connection between each node in each intelligent connected vehicle node set and each node in the remaining intelligent connected vehicle node sets, and establish a composite network of connected vehicle collision avoidance behaviors based on all intelligent connected vehicle state nodes and connection arcs; Based on the cooperative adaptive cruise control system, for any intelligent networked vehicle, the acceleration of the intelligent networked vehicle in the current lane and the acceleration of other intelligent networked vehicles behind the intelligent networked vehicle are obtained before the intelligent networked vehicle performs a feasible driving operation; Obtaining the acceleration of the intelligent networked vehicle traveling in the target lane and the acceleration of other intelligent networked vehicles behind the intelligent networked vehicle after the intelligent networked vehicle performs the feasible driving operation; Set the courtesy factor and benefit factor, and define the attributes of the connection arc to represent the negative benefit between two nodes based on the obtained accelerations, so as to obtain the values ​​of different connection arcs of each intelligent connected vehicle; In the composite network, a decision function is set, and the target function of the target intelligent networked vehicle is established by combining the connection arc. At the same time, to ensure that the current node and the next node of the target intelligent networked vehicle are both selected when the current node and the next node are selected, a constraint condition is set. It is also stipulated that the sum of all decision functions of the target intelligent networked vehicle is 1 each time a connection arc is selected. According to all the connection arcs and nodes in the composite network, different driving routes of the target intelligent networked vehicle when iterating until a preset number of iterations is reached are obtained, and according to the value of each connection arc in each driving route, the value of the sum of all the connection arcs in each driving route is calculated, and by comparing the values ​​of the sum of all the connection arcs, the objective function is solved to obtain the driving route corresponding to the minimum sum of all the connection arcs of the target intelligent networked vehicle when iterating until the preset number of iterations, and the driving route is the optimal driving route of the target intelligent networked vehicle; Based on the process of obtaining the optimal driving route of the above-mentioned target intelligent connected vehicle, the optimal driving routes of all intelligent connected vehicles are obtained in turn, so that all intelligent connected vehicles driving normally on the road behind the faulty intelligent connected vehicle can change lanes in advance according to their corresponding optimal driving routes to avoid the influence range of the preset faulty intelligent connected vehicle.

2. According to claim 1, a method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network, characterized in that: The method of setting the driving operation of each intelligent networked vehicle at the next moment for all the intelligent networked vehicles driving normally on the lanes within the preset range respectively; identifying whether the driving operation of each intelligent networked vehicle at the next moment is feasible based on the cooperative adaptive cruise control system and the safety distance model, and obtaining the feasible driving operation of each intelligent networked vehicle at the next moment includes: According to the preset driving step length and driving direction, the next moment driving operation of each intelligent networked vehicle is set for all the intelligent networked vehicles that are normally driving on the lanes within the preset range; wherein the preset driving step length is +1 or -1 or +2; the preset driving direction is forward, backward, left, and right; Based on the cooperative adaptive cruise control system, for the target intelligent connected vehicle Z i , i = 1, 2, 3..., m, the longitudinal distance between the bumper of the target intelligent networked vehicle and the bumper of the intelligent networked vehicle immediately ahead after the driving operation is recorded as D; the safety distance between the target intelligent networked vehicle and the intelligent networked vehicle immediately ahead after the driving operation is recorded as D safe ; The longitudinal distance between the target intelligent networked vehicle bumper and the bumper of the next intelligent networked vehicle after the target intelligent networked vehicle performs the driving operation is recorded as D back ; The safe distance between the target intelligent networked vehicle and the intelligent networked vehicle immediately behind it after the driving operation is recorded as D back_safe ; Where m represents the number of all intelligent connected vehicles; When D safe <D<2D safe , it is considered that the target intelligent connected vehicle is involved in a collision event; at the same time, it is judged that D back and D back_safe The size relationship; when D back >D back_safe When the speed V of the target intelligent connected vehicle at the next moment is updated by the cooperative adaptive cruise control system rec , and identify the feasible driving operations of the target intelligent connected vehicle at the next moment according to the preset driving step length and driving direction; when D back <D back_safe , the next driving operation set at this time will not be considered; When D>2D safe When , it is considered that the target intelligent networked vehicle is not involved in a collision event, and the target intelligent networked vehicle maintains the current lane, and the speed V of the target intelligent networked vehicle at the next moment is updated through the cooperative adaptive cruise control system. rec , according to the preset +1 or +2 driving step length, identify the feasible driving operation of the target intelligent connected vehicle at the next moment; When D < D safe , the next driving operation set at this time will not be considered; Based on the identification process of the feasible driving operation of the target intelligent connected vehicle at the next moment, whether the driving operation of each intelligent connected vehicle at the next moment is feasible is identified respectively, and the feasible driving operation of each intelligent connected vehicle at the next moment is obtained.

3. According to claim 2, a method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network, characterized in that: The safe distance model includes: During the driving process, when the front vehicle suddenly brakes at the maximum deceleration until the speed reaches 0, the rear vehicle can obtain the information of the front vehicle in real time and brake at the maximum deceleration until the speed reaches 0. At this time, the calculation formula for the minimum safe distance that the two vehicles need to maintain when driving is: D safe =D F -D L +d0 Among them, D F Indicates the distance traveled by the rear vehicle during braking. D L Indicates the distance traveled by the front vehicle during braking. d0 represents the minimum safe distance between two vehicles when the maximum deceleration is 0, and d0 is set to 1.5m; D safe Indicates the minimum safe distance that two vehicles need to maintain when driving. This value is a dynamic value related to the speed of the front and rear vehicles and will change with the driving status of the vehicles; a dcc Indicates the maximum deceleration of the vehicle, set a dcc =4.5m / s.

4. According to claim 2, a method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network, characterized in that: The positions of all the intelligent networked vehicles that are normally traveling on the lanes within the preset range are marked respectively, and recorded as the initial state nodes of the corresponding intelligent networked vehicles; the virtual vehicle positions of each intelligent networked vehicle are obtained according to each feasible driving operation of each intelligent networked vehicle at the next moment, and recorded as the virtual vehicle state nodes of each intelligent networked vehicle; the node set of each intelligent networked vehicle is obtained according to the initial state node of each intelligent networked vehicle and the corresponding virtual vehicle state nodes, and the operation sequence set of each intelligent networked vehicle is obtained, including: Mark the positions of all the intelligent networked vehicles running normally on the lanes within the preset range, and record them as the initial state nodes of the corresponding intelligent networked vehicles, that is, (Z1, 0), (Z2, 0), ..., (Z i ,0),...,(Z m , 0), where m represents the number of all intelligent connected vehicles; Z1, Z2, ..., Z i , ..., Z m represents the number of the intelligent networked vehicle, then the set of all intelligent networked vehicles is H = {Z1, Z2, ..., Z i , ..., Z m }; According to each feasible driving operation of each intelligent connected vehicle at the next moment, each virtual vehicle position of each intelligent connected vehicle is obtained and recorded as the virtual vehicle state node of each intelligent connected vehicle, that is, the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z1 are (Z1, 1), (Z1, 2), (Z1, 3) ... (Z1, a); the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z2 are (Z2, 1), (Z2, 2), (Z2, 3) ... (Z2, b); ...; the virtual vehicle state nodes corresponding to the intelligent connected vehicle Z i The corresponding virtual vehicle state node is (Z i , 1), (Z i , 2), (Z i , 3)...(Z i , n); ...; Intelligent networked vehicle Z m The corresponding virtual vehicle state node is (Z m , 1), (Z m , 2), (Z m , 3)...(Z m , M); According to the initial state nodes of each intelligent connected vehicle and the corresponding virtual vehicle state nodes, the node set of each intelligent connected vehicle is obtained, that is, The node set of the intelligent networked vehicle Z1 is {(Z1, 0), (Z1, 1), (Z1, 2), (Z1, 3) ... (Z1, a)}; the node set of the intelligent networked vehicle Z2 is {(Z2, 0), (Z2, 1), (Z2, 2), (Z2, 3) ... (Z2, b)}; ...; the node set of the intelligent networked vehicle Z i The node set is {(Z i ,0),(Z i , 1), (Z i , 2), (Z i , 3)...(Z i , n)}; ...; Intelligent networked vehicle Z m The node set is {(Z m ,0),(Z m , 1), (Z m , 2), (Z m , 3)...(Z m , M)}; let the set of all intelligent connected vehicle state nodes be N; According to the node set of each intelligent connected vehicle, the operation sequence set of each intelligent connected vehicle is obtained, that is, The operation sequence set of the intelligent connected vehicle Z1 is The operation sequence set of the intelligent connected car Z2 is: ……; Intelligent connected car Z i The set of operation sequences is ……; The set of operation sequences of the intelligent connected vehicle Zm is Where a represents the number of feasible driving operations of the intelligent networked vehicle Z1; b represents the number of feasible driving operations of the intelligent networked vehicle Z2; ...; n represents the number of feasible driving operations of the intelligent networked vehicle Z i The number of feasible driving operations; ...; M represents the number of intelligent connected vehicles Z m The number of possible driving operations.

5. According to the method of claim 1, a multi-vehicle cooperative collision avoidance method for intelligent connected vehicles in a mixed flow environment based on a composite network, characterized in that: The connection arc is used to represent the connection between each node in each intelligent networked vehicle node set and each node in the remaining intelligent networked vehicle node sets. According to all intelligent networked vehicle state nodes and connection arcs, a composite network of networked vehicle collision avoidance behaviors is established, including: The connection arc is used to represent the connection between each node in each intelligent networked vehicle node set and each node in the rest of the intelligent networked vehicle node sets, then Z i The connection arc between each node in the node set and each node in the rest of the intelligent connected vehicle node set is ((WITH i ,0),(Z1,0)),((Z i ,0),(Z1,1)),((Z i ,0),(Z1,2)),...,((Z i ,0),(Z1,a));((Z i ,0),(Z2,0)),((Z i ,0),(Z2,1)),((Z i ,0),(Z2,2)),...,((Z i ,0),(Z2,b));...;((Z i ,0),(From m ,1)),((From i ,0),(From m ,1)),((From i ,0)(From m ,2)),...,((Z i ,0),(From m ,M)); ((WITH i ,1),(Z1,0)),((Z i ,1),(Z1,1)),((Z i ,1),(Z1,2)),...,((Z i ,1),(Z1,a));((Z i ,1),(Z2,0)),((Z i ,1),(Z2,1)),((Z i ,1),(Z2,2)),...,((Z i ,1),(Z2,b));...;((Z i ,1),(From m ,1)),((From i ,1),(From m ,1)),((From i ,1),(From m ,2)),...,((Z i ,1),(From m ,M)); ((WITH i ,2),(Z1,0)),((Z i ,2),(Z1,1)),((Z i ,2),(Z1,2)),...,((Z i ,2),(Z1,a));((Z i ,2),(Z2,0)),((Z i ,2)(Z2,1)),((Z i ,2),(Z2,2)),...,((Z i ,2),(Z2,b));...;((Z i ,2),(From m ,1)),((From i ,2),(From m ,1)),((From i ,2),(From m ,2)),...,((Z i ,2),(From m ,M)); ...； ((Z i ,n),(Z1,0)),((Z i ,n),(Z1,1)),((Z i ,n),(Z1,2)),...,((Z i ,n),(Z1,a));((Z i ,n),(Z2,0)),((Z i ,n),(Z2,1)),((Z i ,n),(Z2,2)),...,((Z i ,n),(Z2,b));...;((Z i ,n),(Z m ,1)),((Z i ,n),(Z m ,1)),((Z i ,n),(Z m ,2)),...,((Z i ,n),(Z m ,M)); Among them, i≠m; According to the above representation of the connection arc of Zi, the connection arc set E of all intelligent connected vehicles is obtained; According to the state node set N and connection arc E of all intelligent connected vehicles, a composite network G (N, E) of connected vehicle collision avoidance behavior is established.

6. The method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network according to claim 5, characterized in that: The attributes defining the connection arc include: According to the negative benefits generated by the lane-changing behavior on the upstream and downstream vehicles in the current lane and the target lane, that is, the driving cost of the intelligent connected vehicle, the properties of the connection arc are defined, and the expression is as follows: Among them, w (v，d)，(u，r) (Z i , O, T) represents the impact of the target intelligent connected vehicle’s operation from the current lane to the target lane on the arc ((v, d), (u, r)) on the target intelligent connected vehicle’s benefit; O and T represent the target intelligent connected vehicle Z i For the current lane and the target lane, if the node after the feasible driving operation is in the same lane as the target intelligent connected vehicle node, O and T are the same; Indicates the acceleration of the target intelligent connected vehicle traveling in the current lane; Indicates the acceleration of the target intelligent connected vehicle traveling in the target lane; Represents the acceleration of the intelligent connected vehicle behind the target intelligent connected vehicle after the feasible driving operation is performed; α T Indicates the acceleration of the intelligent connected vehicle behind the target lane before executing the feasible driving maneuver; Indicates the acceleration of the intelligent connected vehicle behind the current lane after the feasible driving operation is performed; α O The acceleration of the connected vehicle behind the current lane before the feasible driving maneuver is performed; It indicates that the speed advantage obtained by the target intelligent connected vehicle through lane changing has a positive impact on the target intelligent connected vehicle’s revenue; η is the politeness factor; It indicates the impact of executing feasible driving operations on subsequent intelligent connected vehicles to avoid damaging traffic stability; μ is the benefit factor; Indicates the speed advantage of subsequent intelligent connected vehicles in the current lane due to the target intelligent connected vehicle performing feasible driving operations; Through the collaborative adaptive cruise control system, the values ​​of the above-mentioned accelerations can be obtained, thereby obtaining the values ​​of all the connection arcs of Zi; and then obtaining the values ​​of all the connection arcs in the composite network.

7. The method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network according to claim 6, characterized in that: In the composite network, set the decision function, combine the connection arcs, establish the target function of the target intelligent connected vehicle, and set the constraint conditions; According to the preset number of iterations, the minimum value of the objective function is calculated. At this time, the driving route composed of all the connection arcs and nodes is the optimal driving route of the target intelligent connected vehicle, including: In the composite network G(N, E), if the target intelligent connected vehicle is at the node (v, d), there will be a travel cost between the node (v, d) and the next node (u, r) connected to it, that is, the connection arc indicates that there is a negative benefit between the two nodes (v, d) and (u, r); Set the decision function x (v，d)，(u，r) , combined with the connection arc, the objective function of deciding the optimal driving route is established, and its expression is Among them, x (v，d)，(u，r) ∈{0,1}, when x (v，d)，(u，r) Pick 1 When x (v，d)，(u，r) When it is 0, it means that the arc ((v, d), (u, r)) is not selected; E represents the set of arcs; When the connection arc between the front and back nodes is selected, the front and back nodes must be selected. Based on this premise, constraints are set at both ends of the arc to ensure the balance of traffic flow in the composite network. Constraints are set, including: Assuming that the current position of the target intelligent connected vehicle is (v, d), the next position is (u, r), and the constraint condition of the decision variable of node (v, d) is set as Among them, y (v，d) ∈{0,1}, when y (v，d) When 1 is selected, it indicates that the node (v, d) is selected; when y (v，d) When it is 0, it means that the node (v, d) is not selected; N represents the node set; this formula means that when selecting the next node on the (v, d) node, for any selectable node (u, r), y (v，d) Must not be greater than all x (v，d)，(u，r) sum; Assuming that the current position of the target intelligent connected vehicle is (u, r), the previous position is (v, d), and the constraint condition of the decision variable of node (u, r) is set as Among them, y (u，r) ∈{0,1}, when y (u，r) When 1 is selected, it indicates that the node (u, r) is selected; when y (u，r) When it is 0, it means that the node (u, r) is not selected; N represents the node set; this formula means that on the node (u, r), when the previous node is selected, for any selectable node (v, d), y (u，r) Must not be greater than all x (v，d)，(u，r) sum; In the case of multi-vehicle cooperative driving, in order to avoid vehicle conflicts, the sum of the decision functions of each step corresponding to the target vehicle and the intelligent connected vehicle is stipulated to be 1. The constraint condition is expressed as Among them, Z i represents the target intelligent connected vehicle number; H represents the set of all intelligent connected vehicles; f represents the number of feasible driving operations; The target intelligent connected vehicle Z i A set of operation sequences; According to all the connection arcs and nodes in the composite network, different driving routes of the target intelligent connected vehicle are obtained when iterating until a preset number of iterations is reached. According to the value of each connection arc in each driving route, the value of the sum of all the connection arcs in each driving route is calculated. By comparing the values ​​of the sum of all the connection arcs, the objective function is solved to obtain the driving route corresponding to the minimum sum of all the connection arcs of the target intelligent connected vehicle when iterating until the preset number of iterations. At this time, the driving route composed of all the connection arcs and nodes is the optimal driving route of the target intelligent connected vehicle.

8. The method for multi-vehicle coordinated collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network according to claim 7, characterized in that: Set the number of iterations to 200.

9. A multi-vehicle cooperative collision avoidance device for intelligent connected vehicles in a mixed flow environment based on a composite network, characterized in that: include: Memory for storing computer programs; A processor, used to implement the steps of a method for collaborative collision avoidance of multiple intelligent connected vehicles in a mixed flow environment based on a composite network as described in any one of claims 1 to 8 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for multi-vehicle collaborative collision avoidance of intelligent connected vehicles in a mixed flow environment based on a composite network as described in any one of claims 1 to 8.

Images