A traffic flow micro-model optimization method for mixed traffic environment

Abstract

The present invention discloses a traffic flow micro-model optimization method for a mixed traffic environment, including: fusing target vehicle expected attribute information and target vehicle attribute information to obtain vehicle motion potential field strength; using cellular grids to discretize the continuous distribution of motion potential field strengths on each road, and assigning motion potential field strengths to each discrete cellular grid of each road at the current moment; respectively calculating the accelerations of intelligent networked vehicles and manually driven vehicles in a following scenario; optimizing the classic NaSch model according to the vehicle acceleration and the vehicle motion potential field strengths in each road cell after discretization, and obtaining the cellular automaton model evolution rules of the target vehicle under mixed traffic flow. The present invention determines the acceleration of the target vehicle by a comprehensive potential field composed of the velocity potential field of the target vehicle itself and the vehicle potential field of the preceding vehicle, and at the same time determines the random slowing-down rule according to the different vehicle following modes, so as to more accurately characterize the following behavior of vehicles in actual traffic flow.

Description

Technical Field

[0001] The present invention relates to the fields of traffic flow theory, vehicle-road collaboration and intelligent transportation, and in particular to a traffic flow micro-model optimization method for a mixed traffic environment. Background Art

[0002] As an emerging product of the cross-integration of multiple industries such as transportation, automobile industry, artificial manufacturing, and information communication, intelligent networked vehicles have developed rapidly. Equipped with advanced sensing and communication equipment, intelligent networked vehicles can obtain traffic information in real time and use and respond to the acquired traffic information, making up for the shortcomings of traditional traffic systems that information cannot be interacted in real time, which is conducive to targeted and accurate traffic control, improving the operating efficiency and resource utilization of the traffic system, and improving the level of traffic safety.

[0003] The development of intelligent network technology has yet to be perfected and matured. The implementation of intelligent network transportation system will be a task that requires long-term efforts. Before the realization of a complete intelligent network environment, road traffic flow will be a mixed traffic flow composed of intelligent network vehicles and manually driven vehicles for a long time. The difference in traffic information perception ability between intelligent network vehicles and manually driven vehicles leads to differences in their respective motion control, which has different degrees of influence on the evolution of the microscopic state of mixed traffic flow. Studying the evolution process of the microscopic motion state of mixed traffic flow can not only provide the necessary basis for mastering the development law of road traffic situation, but also provide forward-looking direction for the optimization control of mixed traffic flow in the intelligent network environment.

[0004] In the study of traffic flow micro-models, the cellular automaton theory is widely favored, mainly because the theory can simulate complex traffic phenomena more realistically with simple planning and restore the evolution of traffic flow motion state. However, in the current traffic flow micro-model research based on cellular automaton theory, the acceleration and deceleration of vehicles are often set to fixed values, which cannot reflect the behavior of vehicles in the traffic flow environment in response to actual traffic information, and thus it is difficult to restore the micro-state evolution process of real traffic flow; on the other hand, because intelligent connected vehicles in mixed traffic flow have outstanding performance in traffic information perception and suppression of vehicle randomness, traditional cellular automaton models cannot simulate these phenomena well. Summary of the invention

[0005] Purpose of the invention: In view of the above problems, the purpose of the present invention is to provide a traffic flow micro-model optimization method for mixed traffic environments, comprehensively consider the micro-evolution process of mixed traffic flows in intelligent network environments, restore the complex evolution process of the micro-motion state of real traffic flows, and provide a basis for studying the lane changing model of mixed traffic flows.

[0006] Technical solution: A traffic flow micro-model optimization method for a mixed traffic environment of the present invention comprises the following steps:

[0007] Step 1, obtaining the target vehicle's own attribute information and motion attribute information, fusing the target vehicle's expected attribute information and the target vehicle's attribute information to obtain the vehicle's motion potential field strength;

[0008] Step 2: According to the characteristics of the cellular automaton model, the cellular grid is used to discretize the motion potential field strength of each road, and each discrete cellular grid of each road is assigned the motion potential field strength at the current moment;

[0009] Step 3, based on the discretized vehicle motion potential field strength and the motion characteristics of different types of vehicles, respectively calculate the acceleration of the intelligent connected vehicle and the manually driven vehicle in the following scenario;

[0010] Step 4: According to the acceleration of intelligent connected vehicles and manually driven vehicles and the intensity of the vehicle motion potential field in each road cell after discretization, the classic NaSch model is optimized to obtain the cellular automation model evolution rules of the target vehicle from time t to time t+1 under mixed traffic flow.

[0011] Furthermore, the target vehicle's own attribute refers to the mass m of the target vehicle i. i ;

[0012] The motion attribute refers to the speed v generated by the target vehicle i during the motion process. i , acceleration a i , steering angle θ and the distance L between the vehicle i-1 in front;

[0013] The target vehicle expected attribute refers to the expected speed v of the target vehicle i des ;

[0014] Calculate the vehicle potential field strength E of target vehicle i according to the target vehicle attribute information veh The expression is:

[0015]

[0016] Among them, β1 and β2 are parameters that characterize the sensitivity of different types of vehicles to the vehicle motion state;

[0017] Calculate the velocity potential field strength E of the target vehicle based on the target vehicle attribute information and the target vehicle expected attribute information vel The expression is:

[0018] E vel =γ(v i -v des )

[0019] Among them, γ represents the parameter of the feedback degree of different types of vehicles to the speed difference;

[0020] The vehicle motion potential field strength E is obtained based on the target vehicle potential field strength and velocity potential field strength V The expression is:

[0021] E V =E veh +E vel .

[0022] Further, the step 2 comprises:

[0023] The road plane is cut with a certain cell size, and the key node potential field strength value of each cell is obtained. and They represent the potential field strength at the midpoints of the four edges of cell j, so the potential field strength at cell j is:

[0024]

[0025] Further, the step 3 comprises:

[0026] The equivalent mass of target vehicle i is calculated based on its own mass and speed. The expression is:

[0027] M i =m i ×(1.566v i 6.687 ×10 -14 +0.3345)

[0028] According to the information perception capability of intelligent connected vehicles, the acceleration of the target vehicle in the following scenario is calculated under the influence of the potential field force of the preceding vehicle. The expression is:

[0029]

[0030] in, is the field force received by the intelligent connected vehicle target vehicle i in the motion potential field of the preceding vehicle G, is the motion potential field strength at the target vehicle’s location, and are the intensity of the motion potential field at the midpoints of the four edges of the cell where the target vehicle is located;

[0031] According to the information perception ability of the manually driven vehicle, the acceleration of the target vehicle in the following scenario is calculated under the action of the potential field force of the preceding vehicle. The expression is:

[0032]

[0033] in, is the field force that the target vehicle i is subjected to in the motion potential field of the preceding vehicle G, is the motion potential field strength at the target vehicle’s location, and They are respectively the motion potential field strengths at the midpoints of the four edges of the cell where the target vehicle is located.

[0034] Furthermore, the evolution rules of the cellular automaton model in step 4 include the following rules:

[0035] (1) Speed ​​update: v n (t+1)=min(v n (t)+a n ,v max ,LL safe )

[0036] Among them, v n (t) and v n (t+1) is the speed of the nth vehicle at time t and time t+1, v max is the set road speed limit, L is the distance between the target vehicle and the preceding vehicle, L safe The minimum safe distance that the front and rear vehicles should maintain;

[0037] a n is the acceleration of the nth vehicle, when the vehicle type is a manually driven vehicle, When the vehicle type is a smart connected vehicle, When the acceleration calculation result is positive, the target vehicle accelerates; if the calculation result is negative, the target vehicle decelerates;

[0038] (2) Random slowing down:

[0039] Manually driven vehicle: v n (t+1)=max(v n (t+1)-1,0)

[0040] Intelligent connected vehicles: no random slowdowns

[0041] For manually driven vehicles, the random slowing probability is set to p; for intelligent connected vehicles, there is no random slowing phenomenon;

[0042] (3) Position update: x n (t+1)=x n (t)+v n (t+1)

[0043] Among them, x n (t+1) is the position of the nth vehicle at time t+1, xn (t) is the position of the nth vehicle at time t;

[0044] The target vehicle operation is updated according to the above speed update, random slowing down and position update rules, and the update is terminated when the predetermined number of iteration steps is reached.

[0045] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0046] 1. The optimization method of the present invention is oriented to mixed traffic environment. The acceleration of the target vehicle is determined by the comprehensive potential field composed of the velocity potential field of the target vehicle itself and the vehicle potential field of the preceding vehicle. At the same time, the random slowing down rule is determined according to the different following modes of the vehicles. Compared with the NaSch model, the proposed SPF-NS model can more accurately describe the following behavior of vehicles in actual traffic flow.

[0047] 2. The present invention realizes the complex evolution process of the microscopic motion state of mixed traffic flow through simple rules of cellular automata;

[0048] 3. By setting different penetration rates of intelligent connected vehicles, the present invention finds that as the penetration rate of intelligent connected vehicles increases, traffic congestion can be effectively alleviated, and the density value of the time when traffic congestion begins to occur increases, proving the advantages of introducing intelligent connected vehicles into traffic flow;

[0049] 4. This invention is of great significance for future research on the evolution of the microscopic state of mixed traffic flow in complex scenarios, and provides a basis for constructing a multi-lane cellular automaton model of mixed traffic flow based on potential field theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the potential field discretization process;

[0051] Figure 2 A schematic diagram of the vehicle occupying the cell when the cell length is set to 1m and the vehicle length is 5m;

[0052] Figure 3 The spatiotemporal trajectory evolution diagram of traffic flow obtained by simulating the NaSch model when the traffic flow density is 25 vehicles / km and the penetration rate of intelligent connected vehicles is 0%;

[0053] Figure 4 The spatiotemporal trajectory evolution diagram of traffic flow simulated by the SPF-NS model when the penetration rate of intelligent connected vehicles is 0 and the traffic flow density is 25 vehicles / km;

[0054] Figure 5 The spatiotemporal trajectory evolution diagram of traffic flow simulated by the SPF-NS model when the penetration rate of intelligent connected vehicles is 20% and the traffic flow density is 25 vehicles / km;

[0055] Figure 6 The spatiotemporal trajectory evolution diagram of traffic flow simulated by the SPF-NS model when the penetration rate of intelligent connected vehicles is 40% and the traffic flow density is 25 vehicles / km;

[0056] Figure 7 The spatiotemporal trajectory evolution diagram of traffic flow simulated by the SPF-NS model when the penetration rate of intelligent connected vehicles is 60% and the traffic flow density is 25 vehicles / km;

[0057] Figure 8 The spatiotemporal trajectory evolution diagram of traffic flow simulated by the SPF-NS model when the penetration rate of intelligent connected vehicles is 80% and the traffic flow density is 25 vehicles / km;

[0058] Fig. 9 The spatiotemporal trajectory evolution diagram of traffic flow simulated by the SPF-NS model when the penetration rate of intelligent connected vehicles is 100% and the traffic flow density is 25 vehicles / km;

[0059] Fig.10 The flow-density relationship diagram of mixed traffic flow under different penetration rates of intelligent connected vehicles;

[0060] Fig.11 The speed-density relationship diagram of mixed traffic flow under different penetration rates of intelligent connected vehicles;

[0061] Fig.12 This is a graph showing how the congestion ratio changes with traffic flow density under different penetration rates of intelligent connected vehicles. DETAILED DESCRIPTION

[0062] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0063] The traffic flow micro-model optimization method for a mixed traffic environment described in this embodiment includes the following steps:

[0064] Step 1: Obtain the target vehicle's own attribute information and motion attribute information, fuse the target vehicle's expected attribute information and the target vehicle's attribute information, and obtain the vehicle's motion potential field strength.

[0065] The target vehicle's own attribute refers to the mass m of the target vehicle i i ; Motion attribute refers to the speed v generated by the target vehicle i during the motion process i , acceleration a i , steering angle θ and the distance L between the target vehicle and the preceding vehicle i-1; the target vehicle desired attribute refers to the desired speed v of the target vehicle i des .

[0066] Calculate the vehicle potential field strength E of target vehicle i according to the target vehicle attribute information veh The expression is:

[0067]

[0068] Among them, β1 and β2 are parameters that characterize the sensitivity of different types of vehicles to the vehicle motion state;

[0069] Calculate the velocity potential field strength E of the target vehicle based on the target vehicle attribute information and the target vehicle expected attribute information vel The expression is:

[0070] E vel =γ(v i -v des )

[0071] Among them, γ represents the parameter of the feedback degree of different types of vehicles to the speed difference;

[0072] The vehicle motion potential field strength E is obtained based on the target vehicle potential field strength and velocity potential field strength V , the expression is:

[0073] E V =E veh +E vel .

[0074] Step 2: According to the characteristics of the cellular automaton model, the cellular grid is used to discretize the continuous distribution of the motion potential field strength of each road, and the motion potential field strength is assigned to each discrete cellular grid of each road at the current moment, such as Figure 1 The schematic diagram of the potential field discretization process is shown in the figure. The left side represents the potential field strength at the location of the target vehicle, and the right side represents the potential field strength at the location of the preceding vehicle.

[0075] The road plane is cut with a certain cell size, and the key node potential field strength value of each cell is obtained. and They represent the potential field strength at the midpoints of the four edges of cell j, so the potential field strength at cell j is:

[0076]

[0077] Step 3: According to the discretized vehicle motion potential field strength and the motion characteristics of different types of vehicles, the accelerations of the intelligent connected vehicle and the manually driven vehicle in the following scenario are calculated respectively.

[0078] The equivalent mass of target vehicle i is calculated based on its own mass and speed. The expression is:

[0079] Mi =m i ×(1.566v i 6.687 ×10 -14 +0.3345)

[0080] According to the information perception capability of intelligent connected vehicles, the acceleration of the target vehicle in the following scenario is calculated under the influence of the potential field force of the preceding vehicle. The expression is:

[0081]

[0082] in, is the field force received by the intelligent connected vehicle target vehicle i in the motion potential field of the preceding vehicle G, is the motion potential field strength at the target vehicle’s location, and are the intensity of the motion potential field at the midpoints of the four edges of the cell where the target vehicle is located;

[0083] According to the information perception ability of the manually driven vehicle, the acceleration of the target vehicle in the following scenario is calculated under the action of the potential field force of the preceding vehicle. The expression is:

[0084]

[0085] in, is the field force that the target vehicle i is subjected to in the motion potential field of the preceding vehicle G, is the motion potential field strength at the target vehicle’s location, and They are respectively the motion potential field strengths at the midpoints of the four edges of the cell where the target vehicle is located.

[0086] Step 4: According to the acceleration of intelligent connected vehicles and manually driven vehicles and the intensity of the vehicle motion potential field in each road cell after discretization, the classic NaSch model is optimized to obtain the cellular automation model evolution rules of the target vehicle from time t to time t+1 under mixed traffic flow, including the following rules:

[0087] (1) Speed ​​update: v n (t+1)=min(v n (t)+a n ,v max ,LL safe )

[0088] Among them, v n (t) and v n (t+1) is the speed of the nth vehicle at time t and time t+1, v maxis the set road speed limit, L is the distance between the target vehicle and the preceding vehicle, L safe The minimum safe distance that the front and rear vehicles should maintain.

[0089] a n is the acceleration of the nth vehicle, when the vehicle type is a manually driven vehicle, When the vehicle type is a smart connected vehicle, When the acceleration calculation result is positive, the target vehicle accelerates; if the calculation result is negative, the target vehicle decelerates;

[0090] (2) Random slowing down:

[0091] Manually driven vehicle: v n (t+1)=max(v n (t+1)-1,0)

[0092] Intelligent connected vehicles: no random slowdowns

[0093] For manually driven vehicles, the random slowing probability is set to p; for intelligent connected vehicles, there is no random slowing phenomenon;

[0094] (3) Position update: x n (t+1)=x n (t)+v n (t+1)

[0095] Among them, x n (t+1) is the position of the nth vehicle at time t+1, x n (t) is the position of the nth vehicle at time t;

[0096] The target vehicle operation is updated according to the above speed update, random slowing down and position update rules, and the update is terminated when the predetermined number of iteration steps is reached.

[0097] In order to verify that the microscopic model of mixed traffic flow provided by the optimization method of the present invention more effectively describes the following behavior of vehicles in actual mixed traffic flow, the present application is further described in detail below with reference to the accompanying drawings.

[0098] The mixed traffic flow in an artificially set road section is simulated by using the traffic flow model for mixed traffic environment proposed in the present invention, including the following steps:

[0099] Assume that the length of the simulated road is set to 4000m, periodic boundary conditions are used, the lane consists of 4000 cells, each cell is 1m long and 3m wide, the simulation time is 300s, and the simulation accuracy is 1s. Without considering the impact of large vehicles, the vehicle length is set to 5m, that is, a vehicle occupies 5 cells, the width is 2m, the mass is 1000kg, the maximum acceleration is 3m / s2, the maximum deceleration is 5m / s2, the expected speed of the vehicle is 30m / s, and the maximum speed limit on the highway is 35m / s. The schematic diagram of the vehicle occupying the cell is as follows: Figure 2 As shown in Figure 2, the random slowing probability of the manually driven vehicle in the cellular automation is set to 0.3.

[0100] Assume that the traffic flow density is 25 / km and the penetration rate of intelligent connected vehicles is 0. Figure 3 The time-space trajectory evolution diagram of traffic flow obtained by simulating traffic flow using the traditional NaSch model is given. Figure 4 The spatiotemporal trajectory evolution diagram of traffic flow obtained by simulating traffic flow using the SPF-NS model of the present invention is given. Figure 3 and Figure 4 It can be seen that the traffic flow speed under the NaSch model is unstable, vehicles stop and go frequently, and congestion in the traffic flow always exists, which is obviously inconsistent with the actual traffic flow evolution; the traffic flow under the SPF-NS model reflects the process of vehicles adjusting their motion state according to their own motion expectations and the motion restrictions of the preceding vehicles. The traffic flow operation is more stable and can more realistically portray the vehicle following behavior in real situations.

[0101] Figure 5 The spatiotemporal trajectory evolution diagram of traffic flow is given by simulating traffic flow using the SPF-NS model of the present invention when the traffic flow density is 25 vehicles / km and the penetration rate of intelligent connected vehicles is 20%.

[0102] Figure 6 The spatiotemporal trajectory evolution diagram of traffic flow is given by simulating traffic flow using the SPF-NS model when the traffic flow density is 25 vehicles / km and the penetration rate of intelligent connected vehicles is 40%.

[0103] Figure 7 The spatiotemporal trajectory evolution diagram of traffic flow is given by simulating traffic flow using the SPF-NS model when the traffic flow density is 25 vehicles / km and the penetration rate of intelligent connected vehicles is 60%.

[0104] Figure 8 The spatiotemporal trajectory evolution diagram of traffic flow is given by simulating traffic flow using the SPF-NS model when the traffic flow density is 25 vehicles / km and the penetration rate of intelligent connected vehicles is 80%.

[0105] Fig. 9The spatiotemporal trajectory evolution diagram of traffic flow is given when the traffic flow density is 25 vehicles / km and the penetration rate of intelligent connected vehicles is 100% by simulating the traffic flow using the SPF-NS model.

[0106] pass Figure 4-9 It can be seen that when the penetration rate of intelligent connected vehicles is low, due to the high randomness of human-driven vehicles and frequent speed changes, it is easy to cause congestion waves to propagate backwards, and the stability of traffic flow is low. As the penetration rate of intelligent connected vehicles increases, the randomness of vehicle driving behavior decreases, the stability of traffic flow speed increases, the formation and transmission of congestion waves decreases, and the stability of the overall traffic flow is enhanced. When the penetration rate of intelligent connected vehicles reaches 100% and a fully intelligent connected environment is achieved, the stability of traffic flow reaches the highest level, the distance between vehicles tends to be stable, and it has the ability to completely eliminate traffic congestion.

[0107] Fig.10 The relationship between traffic flow and density obtained by simulating traffic flow using the SPF-NS model under different penetration rates of intelligent connected vehicles is given. Fig.10 It can be seen that when the density is lower than the critical density value, the traffic flow increases linearly with the increase of the density value, and when the density is higher than the critical density value, the traffic flow decreases with the increase of the density value until the flow value decreases to 0 and the traffic flow is completely congested. When the penetration rate of intelligent connected vehicles is 40%, the maximum road capacity is 1.3 times that of the pure manual driving environment; when the penetration rate of intelligent connected vehicles is 80%, the maximum road capacity is 1.6 times that of the pure manual driving environment; when the penetration rate of intelligent connected vehicles is 100%, the maximum road capacity is 2.2 times that of the pure manual driving environment. This shows that as the penetration rate of intelligent connected vehicles increases, the road capacity also increases.

[0108] Fig.11 The relationship between traffic flow speed and density obtained by simulating traffic flow using the SPF-NS model under different penetration rates of intelligent connected vehicles is given. Fig.11 It can be seen that the traffic flow speed decreases with the increase of density value until the speed value decreases to 0 and the traffic flow is completely congested. With the increase of the penetration rate of intelligent connected vehicles, the random driving behavior of vehicles in the traffic flow decreases, the vehicle deceleration phenomenon decreases, the traffic flow stability increases, and the average speed of the traffic flow is improved.

[0109] Fig.12 The graph of the congestion ratio and traffic flow density under different penetration rates of intelligent connected vehicles is given. Fig.12It can be seen that when the density value reaches the critical value, the congestion ratio increases with the increase of traffic flow density value. As the penetration rate of intelligent connected vehicles increases, the density value at which traffic flow begins to become congested gradually increases. At the same density value, the congestion ratio of traffic flow decreases with the increase of intelligent connected vehicle penetration rate. This proves the advantage of introducing intelligent connected vehicles in traffic flow.

[0110] In summary, the method of the present invention overcomes the deficiency that the acceleration and deceleration of vehicles in the existing cellular automaton model are fixed values ​​and cannot reflect the adjustment of the vehicle's own movement according to its expected movement restrictions of the preceding vehicle. It provides a traffic flow micro-model optimization method for a mixed traffic environment, optimizes the traditional NaSch model based on potential field theory, effectively describes the reaction of different types of vehicles in the mixed traffic flow to the movement state of the preceding vehicle, and can more accurately simulate the evolution process of the micro-state of the mixed traffic flow.

Claims

1. A traffic flow micro-model optimization method for mixed traffic environment, characterized in that: The following steps are involved: Step 1, obtaining the target vehicle's own attribute information and motion attribute information, fusing the target vehicle's expected attribute information and the target vehicle's attribute information to obtain the vehicle's motion potential field strength; Step 2: According to the characteristics of the cellular automaton model, the cellular grid is used to discretize the motion potential field strength of each road, and each discrete cellular grid of each road is assigned the motion potential field strength at the current moment; Step 3, based on the discretized vehicle motion potential field strength and the motion characteristics of different types of vehicles, respectively calculate the acceleration of the intelligent connected vehicle and the manually driven vehicle in the following scenario; Step 4: According to the acceleration of intelligent connected vehicles and manually driven vehicles and the intensity of the vehicle motion potential field in each road cell after discretization, the classic NaSch model is optimized to obtain the cellular automation model evolution rule of the target vehicle from time t to time t+1 under mixed traffic flow; The target vehicle's own attribute refers to the mass m of the target vehicle i i ; The motion attribute refers to the speed v generated by the target vehicle i during the motion process. i , acceleration a i , steering angle θ and the distance L between the vehicle i-1 in front; The target vehicle expected attribute refers to the expected speed v of the target vehicle i des ; The target vehicle attributes include its own attributes and motion attributes. The vehicle potential field strength E of the target vehicle i is calculated based on the target vehicle attribute information. veh , the expression is: Among them, β1 and β2 represent the sensitivity parameters of different types of vehicles to the vehicle motion state; Calculate the velocity potential field strength E of the target vehicle based on the target vehicle attribute information and the target vehicle expected attribute information vel , the expression is: E vel =γ(v i -v des ) Among them, γ represents the parameter of the feedback degree of different types of vehicles to the speed difference; The vehicle motion potential field strength E is obtained based on the target vehicle potential field strength and velocity potential field strength V The expression is: AND V =And veh +E vel The step 2 comprises: The road plane is cut with a certain cell size, and the key node potential field strength value of each cell is obtained. and They represent the potential field strength at the midpoints of the four edges of cell j, so the potential field strength at cell j is: The step 3 comprises: The equivalent mass of target vehicle i is calculated based on its own mass and speed. The expression is: According to the information perception capability of intelligent connected vehicles, the acceleration of the target vehicle in the following scenario is calculated under the influence of the potential field force of the preceding vehicle. The expression is: in, is the field force received by the intelligent connected vehicle target vehicle i in the motion potential field of the preceding vehicle G, is the motion potential field strength at the target vehicle’s location, and are the motion potential field strengths at the midpoints of the four edges of the cell where the target vehicle is located; According to the information perception ability of the manually driven vehicle, the acceleration of the target vehicle in the following scenario is calculated under the action of the potential field force of the preceding vehicle. The expression is: in, is the field force that the target vehicle i is subjected to in the motion potential field of the preceding vehicle G, is the motion potential field strength at the target vehicle’s location, and are the motion potential field strengths at the midpoints of the four edges of the cell where the target vehicle is located; The evolution rules of the cellular automaton model in step 4 include the following rules: (1) Speed ​​update: v n (t+1)=min(v n (t)+a n ,v max ,LL safe ) Among them, v n (t) and v n (t+1) is the speed of the nth vehicle at time t and time t+1, v max is the set road speed limit, L is the distance between the target vehicle and the preceding vehicle, L safe The minimum safe distance that the front and rear vehicles should maintain; a n is the acceleration of the nth vehicle, when the vehicle type is a manually driven vehicle, When the vehicle type is a smart connected vehicle, When the acceleration calculation result is positive, the target vehicle accelerates; if the calculation result is negative, the target vehicle decelerates; (2) Random slowing down: Manually driven vehicle: v n (t+1)=max(v n (t+1)-1,0) Intelligent connected vehicles: no random slowdowns For manually driven vehicles, the random slowing probability is set to p; for intelligent connected vehicles, there is no random slowing phenomenon; (3) Position update: x n (t+1)=x n (t)+v n (t+1) Among them, x n (t+1) is the position of the nth vehicle at time t+1, x n (t) is the position of the nth vehicle at time t; The target vehicle operation is updated according to the above speed update, random slowing down and position update rules, and the update is terminated when the predetermined number of iteration steps is reached.

Images