A method for constructing a lane-changing model for vehicles in mixed human-machine driving traffic flow
By distinguishing passive and active lane change intention scenarios in human-machine traffic flow, a detailed vehicle lane change model was established, and the problem of insufficient comprehensive existing models was solved, and more accurate analysis of traffic lane change mechanism and traffic congestion were achieved. ease.
Patent Information
- Application Number
- CN202210398315.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-15
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-04-15
AI Technical Summary
The existing model of lane change in human-machine traffic flow vehicles fails to fully consider the impact of vehicle interaction and the collaborative driving behavior of connected autonomous vehicles, resulting in the model being incomplete enough and it is difficult to accurately reflect the lane change mechanism and congestion problems in traffic flow.
A model of lane change for human-machine traffic flow vehicles is proposed, which is divided into passive lane change intention scenarios and active lane change intention scenarios. The lane change model of manual driving vehicles and networked autonomous driving vehicles is established, and the safety distance between vehicles is considered , speed gains and collaborative driving conditions, optimize lane change decisions through mathematical formulas and collaborative driving strategies.
More accurately reflects the lane change mechanism of network-connected autonomous driving vehicles in mixed traffic flows, reflects the interaction effect caused by different vehicle types, explores the impact of lane change control strategies of network-connected autonomous driving vehicles on macroscopic traffic flow characteristics, and provides relief Solutions to traffic congestion.
Smart Images

Figure CN114771522B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle lane-changing simulation, and in particular to a method for constructing a vehicle lane-changing model in a human-machine mixed driving traffic flow. Background Art
[0002] With the development of artificial intelligence, the improvement of Internet of Things technology and the breakthrough of 5G technology, the rapid development of autonomous driving technology is being promoted. Connected and Autonomous Vehicles (CAV) are gradually blending into road traffic, forming a human-machine mixed traffic flow consisting of connected autonomous vehicles and human-driving vehicles (HV). Connected autonomous vehicles are different from manually driven vehicles in terms of vehicle performance, driving behavior, etc. In terms of vehicle performance, connected autonomous vehicles are not affected by psychological and emotional interference, will not drive fatigued, and the reaction time to sudden road situations is much shorter than the reaction time of drivers of manually driven vehicles, which makes them able to accept shorter vehicle gaps than manually driven vehicles in lane changing behavior; at the same time, the cooperative driving characteristics of connected autonomous vehicles make their lane behavior different from that of manually driven vehicles. For example, connected autonomous vehicles can drive in formation through network interconnection and vehicle-to-vehicle communication to maintain a shorter following distance, and communicate with the target lane vehicle to coordinate road space conditions before lane changing occurs. The presence of a queue of connected autonomous vehicles in the target lane will also affect the connected autonomous vehicles' intention to actively change lanes. When driving conditions are not restricted, they will actively change lanes to the target lane and seek platoon driving to reduce the impact of the mixed connection of autonomous vehicles on traffic flow.
[0003] Existing lane-changing models for mixed human-machine traffic flows are often constructed based on different vehicle performance parameters and driving behavior differences, but the following problems still exist: 1) The scenarios considered when constructing vehicle lane-changing models are not comprehensive enough. Existing research on lane-changing models for mixed human-machine traffic flows is mostly limited to the situation where vehicles change lanes due to speed restrictions, and less consideration is given to the active lane-changing of connected autonomous vehicles in order to seek vehicle formations; limited by the complexity of mixed human-machine traffic flows, vehicle queues do not travel in a completely free state, and vehicles in traffic flows will separate and reorganize due to type differences, environmental restrictions, and other reasons; considering the situation where connected autonomous vehicles actively change lanes and merge into vehicle queues, and analyzing the cooperative driving strategies of connected autonomous vehicles, is more meaningful for alleviating mixed traffic congestion and improving traffic capacity. 2) In terms of vehicle interaction, most existing studies consider the impact of connected autonomous vehicles on manually driven vehicles, and less consider the impact of manually driven vehicles on connected autonomous vehicles. Manually driven vehicles in mixed human-machine traffic flow have a certain impact on the established driving behavior of networked autonomous vehicles and make the driving environment of autonomous vehicles more complicated; the construction of the lane change model of networked autonomous vehicles cannot only reflect its performance advantages or driving characteristics, but also consider its role in improving and improving the overall traffic flow; the existing research also lacks macro considerations on the setting of key parameters in the lane change model of networked autonomous vehicles, making it difficult to evaluate the effectiveness and rationality of vehicle decision settings and control strategies. Therefore, considering the interaction between vehicles in mixed human-machine traffic flow, and considering the unique cooperative driving behavior and active lane change intention of networked autonomous vehicles, the lane change conditions for different types of vehicles are calculated separately, which can more accurately reflect the microscopic lane change characteristics of different vehicles in the mixed environment of networked autonomous vehicles, and more deeply reveal the internal mechanism of real traffic, which is of great significance to alleviating the current traffic problems and controlling mixed human-machine traffic flow. Summary of the invention
[0004] In view of the deficiencies in the prior art, the technical problem to be solved by the present invention is to provide a method for constructing a lane-changing model for vehicles in a mixed human-machine driving traffic flow.
[0005] The technical solution adopted by the present invention to solve the above technical problems is as follows:
[0006] A method for constructing a lane-changing model for vehicles in a mixed human-machine driving traffic flow, characterized in that the lane-changing model for vehicles in a mixed human-machine driving traffic flow established by the method includes a lane-changing model for a manually driven vehicle in a passive lane-changing intention scenario, a lane-changing model for a networked automatic driving vehicle, and a lane-changing model for a networked automatic driving vehicle in an active lane-changing intention scenario;
[0007] 1. The lane changing model of a manually driven vehicle in a passive lane changing intention scenario is:
[0008] 1) According to the speed and position information of the vehicle in front of the lane where the lane-changing vehicle is located, it is determined whether the vehicle has a passive lane-changing intention due to speed restriction. When the road conditions cannot meet the vehicle's expected driving speed at the next moment, the vehicle has a lane-changing intention. The speed constraint condition of the lane-changing vehicle is:
[0009] v follow <v expect (1)
[0010] In formula (1), v follow is the speed of the lane-changing vehicle subject to the constraints of the vehicle in front, v expect is the expected speed of the lane-changing vehicle at the next moment;
[0011] 2) After the vehicle has a passive lane change intention, it is determined whether the distance between the lane-changing vehicle and the target lane vehicle meets the safety distance according to formula (2);
[0012] Gap n >Gap safe,n (2)
[0013] In formula (2), Gap n Gap is the distance between the lane-changing vehicle and the target lane vehicle n. safe,n The safe distance required for vehicles changing lanes;
[0014] 3) When the lane-changing vehicle has a passive lane-changing intention and the safety distance is met, the lane-changing vehicle should also consider the speed benefit condition. The speed benefit condition of the lane-changing vehicle is:
[0015] v change >v n (t) (6)
[0016] Among them, v change is the virtual speed of the vehicle after successfully changing lanes, v n (t) is the speed of vehicle n in the target lane;
[0017] Therefore, after the manual driving vehicle has the intention to change lanes passively, it can change lanes freely if the safety distance and speed benefit conditions are met at the same time; if not, it will continue to follow the car;
[0018] 2. The lane-changing model of a connected autonomous driving vehicle in a passive lane-changing intention scenario is:
[0019] 1) After the connected autonomous driving vehicle generates a passive lane change intention, it can change lanes freely if the safety distance, lane change vehicle speed benefit conditions and vehicle overall speed benefit conditions are met at the same time; the speed constraint conditions, safety distance and lane change vehicle speed benefit conditions for the connected autonomous driving vehicle to generate a passive lane change intention meet equations (1), (2) and (6) respectively, and the vehicle overall speed benefit condition is as shown in equation (13):
[0020] v change,rd >γv rd (t) (13)
[0021] Among them, v change,rd is the virtual speed of the vehicle behind the target lane after the vehicle successfully changes lanes, v rd (t) is the speed of the vehicle behind the target lane; γ is the lane-changing decision parameter of the connected autonomous driving vehicle, which indicates the threshold at which the lane-changing behavior of the connected autonomous driving vehicle reduces the speed of the vehicle behind the target lane, and its value range is 0 to 1;
[0022] 2) When a connected autonomous vehicle has a passive lane change intention and the road conditions do not meet the safety distance, the lane-changing vehicle exchanges information with other connected autonomous vehicles in the target lane to determine whether the conditions for collaborative lane change are met. If so, the lane-changing vehicle coordinates the road conditions of the target lane through collaborative driving to meet the safety distance; if not, the lane-changing vehicle continues to follow in its lane;
[0023] There are three scenarios for cooperative driving of vehicles in the target lane, namely: ① The front vehicle in the target lane is a connected autonomous driving vehicle or a connected autonomous driving vehicle queue, and the rear vehicle is a manually driven vehicle. The front vehicle in the target lane cooperates with the vehicle changing lanes to complete the lane change. ② The front vehicle in the target lane is a manually driven vehicle, and the rear vehicle is a connected autonomous driving vehicle or a connected autonomous driving vehicle queue. The rear vehicle in the target lane cooperates with the vehicle changing lanes to complete the lane change. ③ The front vehicle and the rear vehicle in the target lane are both connected autonomous driving vehicles or a connected autonomous driving vehicle queue. The front and rear vehicles are regarded as connected autonomous driving vehicle queues respectively. When safety When the distance satisfies both scenario ① and scenario ②, the connected autonomous driving vehicle platoon is disbanded and coordinated, and the vehicles before the longitudinal position of the lane-changing vehicle are accelerated as coordinated vehicles at the same time, and the coordination process is the same as scenario ①; the vehicles after the longitudinal position of the lane-changing vehicle are decelerated as coordinated vehicles, and the coordination process is the same as scenario ②; when the safety distance satisfies scenario ① but does not satisfy scenario ②, the entire connected autonomous driving vehicle platoon is accelerated as a coordinated vehicle, and the coordination process is the same as scenario ①; when the safety distance does not satisfy scenario ① but satisfies scenario ②, the entire connected autonomous driving vehicle platoon is decelerated as a coordinated vehicle, and the coordination process is the same as scenario ②;
[0024] 3. The lane-changing model of a connected autonomous driving vehicle in the active lane-changing intention scenario is:
[0025] Active lane-changing intention refers to the intention of a connected autonomous vehicle to change lanes when it is affected by a queue of connected autonomous vehicles in the target lane. The lane-changing vehicle identifies the driving status of the vehicle in the current lane. If there is a queue of connected autonomous vehicles ahead and the queue has not reached the maximum queue length, the lane-changing vehicle will keep following the vehicle in the current lane. If there is a manually driven vehicle ahead or a full queue of connected autonomous vehicles, the lane-changing vehicle identifies the driving status of the vehicle in the target lane. If there is a manually driven vehicle ahead or a full queue of connected autonomous vehicles ahead in the target lane, the lane-changing vehicle will keep following the vehicle in the current lane. If there is a queue of connected autonomous vehicles ahead in the target lane and the queue has not reached the maximum queue length, the lane-changing vehicle position is divided into two situations: the lane-changing vehicle is located at the end or the middle of the queue of connected autonomous vehicles.
[0026] When the lane-changing vehicle is at the tail end of a queue of connected autonomous vehicles, that is, the front vehicle in the target lane is a connected autonomous vehicle and the rear vehicle is a manually driven vehicle, the lane-changing process is the same as scenario ①; when the lane-changing vehicle is in the middle of a queue of connected autonomous vehicles, it is necessary to determine the distance between the lane-changing vehicle and the vehicle behind it in the queue. When the distance between the lane-changing vehicle and the vehicle behind it in the queue meets the safety distance, the lane-changing vehicle decelerates and drives to the tail end of the queue of connected autonomous vehicles to change lanes.
[0027] Furthermore, in the manual driving vehicle lane changing model under the passive lane changing intention scenario, the safe distance calculation formula between the lane changing vehicle and the front vehicle in the target lane is:
[0028]
[0029] Among them, Gap safe,fd is the safe distance between the lane-changing vehicle and the vehicle in front of the target lane; v m (t), v fd (t) are the speeds of the lane-changing vehicle m and the vehicle ahead in the target lane fd, b fd is the braking deceleration of the front vehicle fd in the target lane, is the estimated vehicle deceleration, τ m is the lane-changing vehicle reaction time, d is the additional vehicle spacing that the driver of the manual-driven vehicle maintains when facing the connected autonomous-driving vehicle. When the target lane vehicle is also a manual-driven vehicle, d = 0. Otherwise, it is calculated based on the driver’s trust in the connected autonomous-driving vehicle.
[0030] The calculation formula for the safe distance between the lane-changing vehicle and the vehicle behind in the target lane is:
[0031]
[0032] Among them, Gap safe,rd is the safe distance between the lane-changing vehicle and the vehicle behind the target lane, v safe is the safe speed, bm is the braking deceleration of the lane-changing vehicle, x m (t), x rd (t) are the positions of the lane-changing vehicle and the vehicle behind in the target lane, v rd (t) is the speed of the vehicle behind in the target lane, τ rd is the reaction time of the vehicle behind in the target lane.
[0033] Furthermore, in the lane-changing model of a connected autonomous driving vehicle under a passive lane-changing intention scenario, the safe distance calculation formula between the lane-changing vehicle and the vehicle ahead in the target lane is:
[0034]
[0035] The calculation formula for the safe distance between the lane-changing vehicle and the vehicle behind in the target lane is:
[0036]
[0037] Furthermore, the safety distance that cooperative driving needs to meet in scenario ① is:
[0038] Gap fd+1 >Gap safe,fd+1 (15)
[0039] Gap fd+1 Gap is the distance between the front vehicle fd in the target lane and the vehicle fd+1 in the target lane. safe,fd+1 To meet the safety distance of the front vehicle fd in the target lane for cooperative driving;
[0040] Gap fd+1 =x fd+1 (t)-x fd (t)-carL (16)
[0041]
[0042] x fd+1 (t), x fd (t) are the positions of the target lane vehicle fd+1 and the front vehicle fd, respectively. fd+1 (t) is the speed of the vehicle fd+1 in the target lane, b fd+1 is the braking deceleration of the vehicle fd+1 in the target lane, carL is the vehicle length;
[0043] When the safety distance of the target lane is met, the vehicle in front of the target lane adjusts its speed and drives in coordination with the lane-changing vehicle. The state of the vehicle in front of the target lane changes as shown in equations (18) to (20):
[0044]
[0045] vfd (t+Δt)=min[(v fd+1 (t)+a fd (t)Δt,v max ] (19)
[0046]
[0047] Among them, a fd (t) is the acceleration of the vehicle in front of the target lane, T is the expected time interval, v fd (t+Δt) is the speed of the vehicle in front of the target lane at the end of one simulation step cooperative driving, Δt is the simulation step, v max is the maximum speed of the vehicle, x fd (t+Δt) is the position of the front vehicle fd in the target lane at the end of a simulation step cooperative driving, x' fd (t+Δt) is the position of the tail vehicle in the connected autonomous driving vehicle queue at the end of a simulation step cooperative driving, x' fd (t) is the position of the last vehicle in the connected autonomous driving vehicle queue at time t;
[0048] After a simulation step, the safe distance Gap between the lane-changing vehicle and the front vehicle fd in the target lane is fd Update according to formula (21);
[0049] Gap fd =x' fd (t+Δt)-(x m (t)+v m (t)Δt) (21)
[0050] In scenario ②, the safety distance that cooperative driving needs to meet is:
[0051] Gap rd-1 >Gap safe,rd-1 (twenty two)
[0052] Among them, Gap rd-1 Gap is the distance between the target lane rear vehicle rd and the target lane vehicle rd-1. safe,rd-1 To meet the safety distance of the target lane rear vehicle rd cooperative driving;
[0053] Gap rd-1 =x' rd (t)-x rd-1 (t)-carL (23)
[0054]
[0055] x' rd(t) The position of the leading vehicle in the connected autonomous driving vehicle queue, x rd-1 (t) is the position of the vehicle rd-1 in the target lane, v rd-1 (t) is the speed of the vehicle rd-1 in the target lane, τ rd-1 is the reaction time of vehicle rd-1 in the target lane, b rd-1 , b rd are the braking decelerations of the target lane vehicle rd-1 and the target lane vehicle rd respectively;
[0056] When the target lane safety distance is met, the vehicle behind the target lane adjusts its own speed and drives in coordination with the lane-changing vehicle. The state of the vehicle behind the target lane changes as shown in equations (25) to (27):
[0057]
[0058] v rd (t+Δt)=max[(v rd (t)+a rd (t)Δt,0] (26)
[0059]
[0060] Among them, a rd (t) is the acceleration of the vehicle behind the target lane, v rd (t+Δt), x rd (t+Δt) is the speed and position of the vehicle behind in the target lane at the end of a simulation step cooperative driving process, x' rd (t+Δt) is the position of the tail vehicle in the connected autonomous driving vehicle queue at the end of a simulation step cooperative driving process;
[0061] Gap m =x m (t)+v m (t)Δt-x rd (t+Δt) (28)
[0062] After a simulation step, the safe distance Gap between the lane-changing vehicle and the vehicle behind in the target lane m Update according to formula (28).
[0063] Compared with the prior art, the present invention has the following beneficial effects:
[0064] (1) The present invention fully considers the differences in vehicle driving states and the diversity of vehicle driving behaviors under the conditions of mixed human-machine driving traffic flow, especially divides the lane changing behaviors of vehicles in mixed traffic flow into passive lane changing intention scenarios and active lane changing intention scenarios, and calculates the safety gaps between manually driven vehicles and networked autonomous driving vehicles for the passive lane changing intention scenarios, distinguishes between free lane changing scenarios when the lane changing space conditions are met and networked autonomous driving vehicle collaborative lane changing scenarios when the space conditions are limited; constructs active lane changing intention scenarios for the networked autonomous driving vehicle platooning behaviors; can more accurately reflect the lane changing mechanism of networked autonomous driving vehicles in mixed human-machine driving traffic flow, and at the same time reflect the interactive effects caused by different vehicle types in mixed human-machine driving traffic flow.
[0065] (2) The lane-changing model in the present invention takes into account the differences in decision-making of different types of vehicles. In the free lane-changing scenario, the manually driven vehicle uses speed benefit as the lane-changing decision factor, while the networked autonomous driving vehicle uses its own speed benefit and the impact on other vehicles in the traffic flow as lane-changing decision factors. This can reflect the consideration and contribution of the networked autonomous driving vehicle to the overall traffic flow benefit, explore the impact trend of the lane-changing control strategy of the networked autonomous driving vehicle on the macro traffic flow characteristics, and provide solutions to alleviate the current traffic congestion and other problems in the traffic flow. At the same time, it has guiding significance for the control of the future networked autonomous driving vehicles mixed into the road traffic flow. BRIEF DESCRIPTION OF THE DRAWINGS
[0066] Figure 1 It is the overall flow chart of the present invention;
[0067] Figure 2 is a schematic diagram of a manually driven vehicle changing lanes freely under a passive lane change intention scenario;
[0068] Figure 3 It is a schematic diagram of the free lane change of a connected autonomous driving vehicle in a passive lane change intention scenario;
[0069] Figure 4 This is a schematic diagram of the coordinated lane change of the vehicle in front of the target lane in the coordinated lane change scenario;
[0070] Figure 5 This is a schematic diagram of the coordinated lane change of the vehicle behind the target lane in the coordinated lane change scenario;
[0071] Figure 6 It is a schematic diagram of the coordinated lane changing of vehicle queues in the coordinated lane changing scenario;
[0072] Figure 7 This is a schematic diagram of the coordinated lane changing of a vehicle queue in the coordinated lane changing scenario;
[0073] Figure 8 This is a schematic diagram of the coordinated lane change of a vehicle queue during deceleration in a coordinated lane change scenario;
[0074] Fig. 9 This is a schematic diagram of a vehicle changing lanes at the end of a queue in an active lane change scenario;
[0075] Fig.10 This is a schematic diagram of a vehicle changing lanes in the middle of a queue in an active lane change scenario;
[0076] Fig.11 It is the Matlab simulation flow chart;
[0077] Fig.12 It is a curve diagram of the change of vehicle density and the speed difference between the front and rear vehicles when γ = 0.8;
[0078] Fig.13 is a curve diagram of the penetration rate of connected autonomous driving vehicles and the maximum speed difference between the front and rear vehicles corresponding to different values of the lane change decision parameter γ of the connected autonomous driving vehicle;
[0079] Figure 14(a) is a graph showing the density and speed of vehicles in a mixed human-machine driving traffic flow at different penetration rates of connected autonomous vehicles;
[0080] Figure 14(b) is a graph of vehicle density and flow in mixed human-machine driving traffic flow at different penetration rates of connected autonomous vehicles;
[0081] Fig.15 It is the overall lane-changing frequency diagram of vehicles in mixed human-machine driving traffic flow;
[0082] Fig.16 It is the frequency diagram of coordinated lane changing of connected autonomous vehicles in mixed human-machine driving traffic flow;
[0083] Figure 17(a) is a spatiotemporal distribution trajectory diagram of vehicles in a mixed human-machine driving traffic flow when the penetration rate of connected autonomous driving vehicles is 0;
[0084] Figure 17(b) is a spatiotemporal distribution trajectory diagram of vehicles in a mixed human-machine driving traffic flow when the penetration rate of connected autonomous vehicles is 0.3;
[0085] Figure 17(c) is a spatiotemporal distribution trajectory diagram of vehicles in a mixed human-machine driving traffic flow when the penetration rate of connected autonomous vehicles is 0.7;
[0086] Figure 17(d) is a spatiotemporal distribution trajectory diagram of vehicles in a mixed human-machine driving traffic flow when the penetration rate of connected autonomous vehicles is 1;
[0087] Fig.18 It is a curve chart of the penetration rate of connected autonomous vehicles and the average value of vehicle speed fluctuation at different vehicle speeds;
[0088] Fig.19 It is the longitudinal collision risk map of traffic flow. DETAILED DESCRIPTION
[0089] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods, but the protection scope of the present application is not limited thereto.
[0090] like Figure 1 As shown, the present invention is a method for constructing a lane-changing model for vehicles in a mixed human-machine driving traffic flow (hereinafter referred to as the method). The lane-changing model for vehicles in a mixed human-machine driving traffic flow established by the method includes a lane-changing model for a manually driven vehicle in a passive lane-changing intention scenario, a lane-changing model for a networked autonomous driving vehicle, and a lane-changing model for a networked autonomous driving vehicle in an active lane-changing intention scenario;
[0091] 1. The lane changing model of a manually driven vehicle in a passive lane changing intention scenario is as follows:
[0092] 1) According to the speed and position information of the vehicle in front of the lane where the lane-changing vehicle is located, determine whether the vehicle has a passive lane-changing intention due to speed restriction; passive lane-changing intention refers to the lane-changing intention generated by the vehicle in pursuit of a higher driving speed when the speed of the vehicle in front is slow, that is, when the road conditions cannot meet the vehicle's expected driving speed at the next moment, the vehicle has a lane-changing intention. The speed constraint of the lane-changing vehicle is:
[0093] v follow <v expect (1)
[0094] In formula (1), v follow is the speed of the lane-changing vehicle subject to the constraints of the vehicle in front, v expect is the expected speed of the lane-changing vehicle at the next moment;
[0095] 2) After the vehicle has a passive lane change intention, it is determined whether the distance between the lane-changing vehicle and the target lane vehicle meets the safety distance according to formula (2);
[0096] Gap n >Gap safe,n (2)
[0097] In formula (2), Gap n Gap is the distance between the lane-changing vehicle and the target lane vehicle n. safe,n is the safety distance required for lane-changing vehicles, n is fd or rd, fd represents the front vehicle in the target lane, and rd represents the rear vehicle in the target lane;
[0098] When n is fd, the safe distance between the lane-changing vehicle and the vehicle in front of the target lane is calculated as follows:
[0099]
[0100] Among them, Gap safe,fd v is the safe distance that the lane-changing vehicle m needs to maintain with the vehicle fd in front of the target lane; m (t), vfd (t) are the speeds of the lane-changing vehicle m and the vehicle ahead in the target lane fd, b fd is the braking deceleration of the front vehicle fd in the target lane, b fd =-3.3m·s -2 ; is the estimated vehicle deceleration, τ m is the lane-changing vehicle reaction time. When the lane-changing vehicle is a manual driving vehicle, τ m =1.5s; d is the additional vehicle distance that the driver of the manually driven vehicle maintains when facing the connected autonomous driving vehicle. When the vehicle in the target lane is also a manually driven vehicle, d=0. Otherwise, it is calculated based on the driver’s trust in the connected autonomous driving vehicle.
[0101] When n is rd, the safe distance that the lane-changing vehicle needs to maintain with the vehicle behind the target lane is calculated. The lane-changing vehicle does not make emergency braking. The lane-changing vehicle determines the safe distance required for lane changing based on the speed and position of the vehicle behind the target lane. Figure 2 The safe speed v required for the lane-changing vehicle to maintain a safe distance with the vehicle behind in the target lane is obtained by using the vehicle position relationship in the target lane. safe for:
[0102]
[0103] The safe distance between the lane-changing vehicle and the vehicle behind in the target lane is calculated based on the safe speed as shown in formula (5):
[0104]
[0105] Among them, Gap safe,rd is the safe distance that the lane-changing vehicle m needs to maintain with the vehicle behind the target lane rd, b m is the braking deceleration of the lane-changing vehicle m, x m (t), x rd (t) represent the positions of the lane-changing vehicle m and the vehicle behind the target lane rd, respectively, and v rd (t) is the speed of the vehicle rd behind the target lane, τ rd is the reaction time of the vehicle rd behind the target lane, carL is the vehicle length, which is 5m;
[0106] 3) When the lane-changing vehicle has a passive lane-changing intention and the safety distance is met, the lane-changing vehicle should also consider the speed benefit condition. The speed benefit condition of the lane-changing vehicle is:
[0107] v change >v n (t) (6)
[0108] Among them, v changeis the virtual speed of the vehicle after successfully changing lanes, which is calculated by the vehicle following model according to the type of vehicle in front of the target lane; v n (t) is the speed of vehicle n in the target lane;
[0109] Therefore, after a manually driven vehicle has the intention to passively change lanes, it can change lanes freely if both the safety distance and speed benefit conditions are met; if not, it will continue to follow the car.
[0110] 2. The lane-changing model of a connected autonomous driving vehicle in a passive lane-changing intention scenario is:
[0111] 1) After the connected autonomous driving vehicle generates a passive lane change intention, it can change lanes freely if the safety distance, lane change vehicle speed benefit conditions and vehicle overall speed benefit conditions are met at the same time; the speed constraint conditions, safety distance and lane change vehicle speed benefit conditions for the connected autonomous driving vehicle to generate a passive lane change intention meet equations (1), (2) and (6) respectively;
[0112] In the free lane-changing behavior of connected autonomous vehicles, the vehicle reaction time is negligible. Figure 3 The safe distance between the lane-changing vehicle and the front vehicle fd in the target lane is obtained from the vehicle position relationship in the target lane as shown in equations (7) to (9):
[0113] Gap safe,fd =x fd (t0)-x m (t0)-carL=Δx m -Δx fd (7)
[0114] Δx fd =x fd (t1)-x fd (t0) = -v fd (t0) 2 / 2b fd (8)
[0115]
[0116] Among them, x m (t0), x fd (t0) are the positions of the lane-changing vehicle m and the vehicle ahead in the target lane fd at the initial time of lane changing, Δx m , Δx fd are the position changes of the front vehicle fd and the lane-changing vehicle m in the target lane during the lane-changing process, and x m (t1), x fd (t1) are the positions of the lane-changing vehicle m and the vehicle ahead in the target lane fd at the end of the lane change;
[0117] It can be deduced that the safe distance between the lane-changing vehicle and the vehicle in front of the target lane needs to be maintained as shown in formula (10):
[0118]
[0119] When calculating the safe distance that the lane-changing vehicle needs to maintain with the vehicle behind the target lane, the lane-changing vehicle does not make an emergency brake. The lane-changing vehicle determines the required safe distance for lane changing based on the speed and position of the vehicle behind the target lane. The safe speed required for the lane-changing vehicle to maintain a safe distance with the vehicle behind the target lane is as shown in formula (11):
[0120]
[0121] The safe distance is calculated from the safe speed as shown in formula (12):
[0122]
[0123] In addition, in addition to considering its own speed benefit, the lane-changing decision-making process of the connected autonomous driving vehicle should also consider the impact of the lane-changing behavior on other vehicles in the traffic flow, and use the speed loss caused by the lane-changing behavior to the vehicle behind the target lane as the standard to determine whether to change lanes; if the lane-changing behavior of the connected autonomous driving vehicle causes the speed of the vehicle behind the target lane to decrease by more than a certain range, the vehicle cancels the lane change, that is, the connected autonomous driving vehicle needs to change lanes only when the running speed of the vehicle behind the target lane is guaranteed. The overall speed benefit condition of the vehicle is as follows:
[0124] v change,rd >γv rd (t) (13)
[0125] Among them, v change,rd v is the virtual speed of the vehicle behind the target lane after the vehicle successfully changes lanes, which is calculated by different following models according to different vehicle types; rd (t) is the speed of the vehicle behind the target lane at time t; γ is the lane-changing decision parameter of the connected autonomous driving vehicle, which indicates the threshold at which the lane-changing behavior of the connected autonomous driving vehicle reduces the speed of the vehicle behind the target lane. The value range is 0 to 1. The value of the lane-changing parameter γ will directly affect the lane-changing decision of the connected autonomous driving vehicle. The closer γ is to 1, the more the connected autonomous driving vehicle is concerned about the interference caused by its lane-changing behavior to the vehicle behind the target lane. Its value can be defined based on simulation analysis and is optimally determined by the stability of the macroscopic traffic flow under different lane-changing parameters.
[0126] The stability of traffic flow is expressed by counting the average value of the speed difference between the front and rear vehicles in the traffic flow. The calculation formula is as follows:
[0127]
[0128] Where i is the vehicle number; j is the lane number; vj,i 、v j,i+1 are the speeds of the i-th vehicle and the i+1-th vehicle in lane j respectively; N maxj is the total number of vehicles in lane j in the traffic flow;
[0129] 2) Since the IoT autonomous driving vehicle has the communication function, when the connected autonomous driving vehicle has the intention to passively change lanes and the road conditions do not meet the safety interval, the lane-changing vehicle can exchange information with other connected autonomous driving vehicles in the target lane to determine whether the conditions for collaborative lane change are met. If so, the road conditions of the target lane are coordinated through collaborative driving to make the road conditions meet the safety interval; if not, the lane-changing vehicle continues to follow in the lane; information exchange only occurs when the lane-changing vehicle is a connected autonomous driving vehicle and there is a connected autonomous driving vehicle in the target lane at the same time;
[0130] There are three scenarios for cooperative driving of vehicles in the target lane: ① The front vehicle in the target lane is a connected autonomous driving vehicle or a connected autonomous driving vehicle queue, and the rear vehicle is a manually driven vehicle; ② The front vehicle in the target lane is a manually driven vehicle, and the rear vehicle is a connected autonomous driving vehicle or a connected autonomous driving vehicle queue; ③ Both the target lane vehicle and the rear vehicle are connected autonomous driving vehicles or a connected autonomous driving vehicle queue. Whether the vehicle can successfully complete the cooperative driving behavior depends not only on the lane-changing vehicle and the vehicles participating in the cooperative driving in the target lane, but also on the spatial conditions between the vehicles participating in the cooperative driving and the vehicles in front and behind them. Based on this, the three scenarios are calculated separately:
[0131] ① When the front vehicle in the target lane is a connected autonomous vehicle or a connected autonomous vehicle platoon, and the rear vehicle is a manually driven vehicle, the front vehicle in the target lane cooperates with the vehicle changing lanes to complete the lane change (see Figure 4 ); The safety distance that needs to be met for cooperative driving is as shown in formula (15). In the process of calculating the safety distance, the connected autonomous driving vehicle team is regarded as one vehicle. Therefore, in formulas (15) to (20), when the front vehicle in the target lane is driving cooperatively, fd represents the front vehicle in the target lane, and when the connected autonomous driving vehicle team is driving cooperatively, fd represents the connected autonomous driving vehicle team;
[0132] Gap fd+1 >Gap safe,fd+1 (15)
[0133] Among them, Gap fd+1 Gap is the distance between the front vehicle fd in the target lane and the vehicle fd+1 in the target lane. safe,fd+1 To meet the safety distance of the front vehicle fd cooperative driving in the target lane;
[0134] Gap fd+1 =x fd+1 (t)-xfd (t)-carL (16)
[0135]
[0136] Among them, x fd+1 (t), x fd (t) are the positions of the target lane vehicle fd+1 and the front vehicle fd, respectively. fd+1 (t) is the speed of the target lane vehicle fd+1 at time t, b fd+1 is the braking deceleration of the vehicle fd+1 in the target lane;
[0137] When the target lane safety distance is met, the vehicle fd in front of the target lane adjusts its own speed and drives in coordination with the lane-changing vehicle m. The state of the vehicle fd in front of the target lane changes as shown in equations (18) to (20):
[0138]
[0139] v fd (t+Δt)=min[(v fd+1 (t)+a fd (t)Δt,v max ] (19)
[0140]
[0141] Among them, a fd (t) is the acceleration of the front vehicle fd in the target lane, T is the expected time interval, v fd (t+Δt) is the speed of the front vehicle fd in the target lane at the end of one simulation step cooperative driving, Δt is the simulation step, v max is the maximum speed of the vehicle, x fd (t+Δt) is the position of the front vehicle fd in the target lane at the end of a simulation step cooperative driving, x' fd (t+Δt) is the position of the tail vehicle in the connected autonomous driving vehicle queue at the end of a simulation step cooperative driving, x' fd (t) is the position of the last vehicle in the connected autonomous driving vehicle queue at time t;
[0142] After a simulation step, the safety distance between the lane-changing vehicle m and the front vehicle fd in the target lane is updated according to formula (21);
[0143] Gap fd =x' fd (t+Δt)-(x m (t)+v m (t)Δt) (21)
[0144] ② When the front vehicle in the target lane is a manually driven vehicle and the rear vehicle is a connected autonomous vehicle or a platoon of connected autonomous vehicles, the rear vehicle in the target lane cooperates with the vehicle to complete the lane change (see Figure 5 ); The safety distance that needs to be met for cooperative driving is as shown in formula (22). The connected autonomous driving vehicle team is regarded as one vehicle. Therefore, in formulas (22) to (27), when the vehicle behind the target lane is driving cooperatively, rd represents the vehicle behind the target lane. When the connected autonomous driving vehicle team is driving cooperatively, rd represents the connected autonomous driving vehicle team.
[0145] Gap rd-1 >Gap safe,rd-1 (twenty two)
[0146] Among them, Gap rd-1 Gap is the distance between the target lane rear vehicle rd and the target lane vehicle rd-1. safe,rd-1 To meet the safety distance of the rear vehicle in the target lane for cooperative driving;
[0147] Gap rd-1 =x' rd (t)-x rd-1 (t)-carL (23)
[0148]
[0149] Among them, x' rd (t) The position of the leading vehicle in the connected autonomous driving vehicle queue, x rd-1 (t) is the position of the vehicle rd-1 in the target lane, v rd-1 (t) is the speed of the vehicle rd-1 in the target lane, τ rd-1 is the reaction time of vehicle rd-1 in the target lane, b rd-1 , b rd are the braking decelerations of the target lane vehicle rd-1 and the target lane vehicle rd respectively;
[0150] When the target lane safety distance is met, the target lane rear vehicle rd adjusts its own speed and drives in coordination with the lane-changing vehicle m. The state of the target lane rear vehicle rd changes as shown in equations (25) to (27):
[0151]
[0152] v rd (t+Δt)=max[(v rd (t)+a rd (t)Δt,0] (26)
[0153]
[0154] Among them, a rd (t) is the acceleration of the vehicle rd in the target lane, v rd (t+Δt), x rd (t+Δt) is the speed and position of the vehicle behind the target lane at the end of a simulation step cooperative driving process, x' rd (t+Δt) is the position of the tail vehicle in the connected autonomous driving vehicle queue at the end of a simulation step cooperative driving process;
[0155] After a simulation step, the safety distance between the lane-changing vehicle m and the vehicle behind the target lane rd is updated according to formula (28);
[0156] Gap m =x m (t)+v m (t)Δt-x rd (t+Δt) (28)
[0157] ③ When both the front and rear vehicles in the target lane are connected autonomous vehicles or a connected autonomous vehicle fleet, the front and rear vehicles in the target lane are connected autonomous vehicles. The front and rear vehicles can be regarded as connected autonomous vehicle fleets respectively. There are three situations for the coordinated driving behaviors of connected autonomous vehicles or connected autonomous vehicle fleets, namely: i. Disbanding and coordination of connected autonomous vehicle fleets (see Figure 6 );ii. Connected autonomous driving vehicle platoon acceleration coordination (see Figure 7 );iii. Coordinated deceleration of connected autonomous driving vehicle platoons (see Figure 8 ); specifically as follows:
[0158] When the safety distance i satisfies both scenario ① and scenario ②, the connected autonomous driving vehicle queue is disbanded and coordinated, and the vehicles before the longitudinal position of the lane-changing vehicle m are accelerated as cooperative vehicles at the same time, and the coordination process is the same as scenario ①; the vehicles after the longitudinal position of the lane-changing vehicle m are decelerated as cooperative vehicles, and the coordination process is the same as scenario ②;
[0159] ii. When the safety distance meets scenario ① but does not meet scenario ②, the entire connected autonomous driving vehicle fleet is accelerated as a cooperative vehicle and driven in coordination with the lane-changing vehicle m. The coordination process is the same as scenario ①.
[0160] iii. When the safety distance does not meet scenario ① but meets scenario ②, the entire connected autonomous driving vehicle fleet is treated as a cooperative vehicle and decelerated to drive in coordination with the lane-changing vehicle m. The coordination process is the same as scenario ②.
[0161] At each simulation step, the safe distance between the lane-changing vehicle and the cooperative vehicle needs to be re-judged. If the distance is met, the cooperative driving lane-changing behavior is executed. If not, the cooperative driving lane-changing behavior is re-judged and executed at the next simulation step.
[0162] 3. The lane-changing model of the connected autonomous driving vehicle in the active lane-changing intention scenario is as follows:
[0163] Active lane-changing intention refers to the intention of a connected autonomous vehicle to change lanes when it is affected by a queue of connected autonomous vehicles in the target lane. The lane-changing vehicle identifies the driving status of the vehicle in the current lane. If there is a queue of connected autonomous vehicles ahead and the queue has not reached the maximum queue length, the lane-changing vehicle will keep following the vehicle in the current lane. If there is a manually driven vehicle ahead or a full queue of connected autonomous vehicles, the lane-changing vehicle identifies the driving status of the vehicle in the target lane. If there is a manually driven vehicle ahead or a full queue of connected autonomous vehicles ahead in the target lane, the lane-changing vehicle will keep following the vehicle in the current lane. If there is a queue of connected autonomous vehicles ahead in the target lane and the queue has not reached the maximum queue length, the lane-changing vehicle position is divided into two situations: the lane-changing vehicle is located at the end or the middle of the queue of connected autonomous vehicles.
[0164] When the lane-changing vehicle is at the tail end of the connected autonomous vehicle queue, that is, the front vehicle in the target lane is a connected autonomous vehicle and the rear vehicle is a manually driven vehicle, the lane-changing process is the same as scenario ① (see Fig. 9 );
[0165] When the lane-changing vehicle is in the middle of the connected autonomous vehicle queue, it is necessary to determine the distance between the lane-changing vehicle m and the vehicle r behind it in the connected autonomous vehicle queue. When the distance between the lane-changing vehicle m and the vehicle r behind it in the connected autonomous vehicle queue meets the safety distance, the lane-changing vehicle decelerates and drives to the end of the connected autonomous vehicle queue to change lanes and merge into the queue (see Fig.10 ); In this process, the safe distance for lane-changing vehicle m to decelerate is as shown in formula (29):
[0166] Gap r >Gap safe,r (29)
[0167] Among them, Gap r Gap is the distance between the lane-changing vehicle m and the front of the vehicle r behind it in the connected autonomous driving vehicle queue. safe,r To meet the safety distance for lane-changing vehicle m to decelerate;
[0168] Gap r =x m (t)-x r (t) (30)
[0169] Gap safe,r =v r (t)τ r -v r (t) 2 / 2b r +v m(t) 2 / 2b m (31)
[0170] Among them, x r (t), v r (t) is the position and speed of the rear vehicle r in the connected autonomous vehicle queue, τ r , b r are the reaction time and braking deceleration of the rear vehicle r in the connected autonomous vehicle queue;
[0171] The state changes of the lane-changing vehicle m when decelerating are shown in equations (32) to (33):
[0172] v m (t+Δt)=max[v m (t)+b m Δt,0] (32)
[0173]
[0174] Among them, v m (t+Δt), x m (t+Δt) are the speed and position of the lane-changing vehicle m at the end of a simulation step;
[0175] In the next simulation step, the lane-changing vehicle m needs to re-judge the safe distance and slow down until its longitudinal position is at the tail end of the connected autonomous driving vehicle queue in the target lane. Then the lane-changing vehicle m can change lanes to the target lane.
[0176] The mixed human-machine driving traffic flow includes five vehicle following models: HV-HV, CAV-HV, HV-CAV, CAV-CAV, and CAVplat-CAVplat. HV represents manually driven vehicles, CAV represents connected autonomous driving vehicles, and CAVplat represents a queue of connected autonomous driving vehicles.
[0177] The traffic flow is simulated using the MTALAB software using the human-machine mixed driving traffic flow vehicle lane changing model of the present invention. The simulation flow chart is shown in Fig.11 ; First, generate the initial simulation environment, with a simulation cell length of 0.5m; the road environment is a periodic one-way two-lane highway with a total length of 6000m, a road speed limit of 120km / h, and free lane change between lanes; generate the initial vehicle distribution, with a vehicle length of 5m and an initial speed of 0; the vehicle type and initial position are randomly generated, and the driver characteristics of manually driven vehicles are randomly generated;
[0178] Then, the simulation time-step cycle is started, with a total simulation time of 2000s and a simulation step of 1s. In each simulation step, the tail vehicle in the connected autonomous driving vehicle queue is identified, all vehicles are traversed and the expected driving speed of the vehicle is calculated. The passive lane-changing intention and passive lane-changing conditions of the vehicle are determined, and if the lane-changing conditions are met, the vehicle performs the lane-changing. For connected autonomous driving vehicles that do not meet the free lane-changing safety interval, the cooperative lane-changing safety interval is determined. If it is met, the cooperative driving assists the vehicle to perform the lane-changing, otherwise it keeps following the vehicle.
[0179] Traverse all connected autonomous vehicles, determine the active lane change intention and active lane change conditions of connected autonomous vehicles, and for vehicles with active lane change intention, if the active lane change conditions are met, execute lane change and merge into the connected autonomous vehicle queue of the target lane; if not, keep driving in the current lane;
[0180] Traverse the vehicles and execute vehicle status updates, clear the lane-changing vehicles in this lane, and insert lane-changing vehicles in other lanes according to the vehicle head position; after the simulation time step ends the loop, the following is obtained: Figures 12 to 19 The simulation results are shown in Figure 2.
[0181] Determine the lane-changing decision parameter γ of the connected autonomous driving vehicle through simulation. Fig.12 The curve diagram of the change of vehicle density and the speed difference between the front and rear vehicles when γ=0.8 shows that the speed difference between the front and rear vehicles in the traffic flow increases first and then decreases as the vehicle density increases. The rising stage represents that the stability of the traffic flow gradually deteriorates, and the maximum value of the speed difference between the front and rear vehicles represents the worst traffic flow stability. As the vehicle density continues to increase, the speed difference between the front and rear vehicles gradually decreases. This is because the absolute value of the average speed of the traffic flow decreases due to the increase in the number of vehicles, and the speed difference between the front and rear vehicles decreases accordingly. At the same time, under different penetration rates of connected autonomous driving vehicles, the corresponding vehicle density when the maximum speed difference between the front and rear vehicles occurs is also different. In order to comprehensively compare the impact of the reduction ratio γ on the stability of traffic flow under different states, the maximum speed difference between the front and rear vehicles in the traffic flow under different penetration rates of connected autonomous driving vehicles is selected as a reference for determining the value of the lane change decision parameter γ for connected autonomous driving vehicles.
[0182] Fig.13 is a curve diagram of the penetration rate of connected autonomous driving vehicles and the maximum speed difference between the front and rear vehicles corresponding to different values of the lane change decision parameter γ of the connected autonomous driving vehicle, Fig.13It can be seen that the maximum speed difference between the front and rear vehicles gradually decreases with the increase of the penetration rate of connected autonomous vehicles. A comprehensive comparison of the maximum speed difference between the front and rear vehicles under different penetration rates of connected autonomous vehicles shows that when γ = 0.8, the speed difference between the front and rear vehicles is relatively minimum, and the traffic flow is in a relatively stable state. Therefore, it is determined that the reduction ratio of the speed of the vehicle behind the target lane when the connected autonomous vehicle changes lanes is γ = 0.8, that is, the connected autonomous vehicle only performs lane change behavior when the vehicle behind the target lane is driving at 80% of its expected speed.
[0183] Through simulation, starting from the aspects of traffic flow operation efficiency, stability, safety, etc., we selected indicators such as average traffic speed, flow, vehicle lane change frequency, vehicle spatiotemporal trajectory, vehicle speed fluctuation, collision risk, etc. to analyze the impact of the mixing of connected autonomous driving vehicles on the macro traffic flow characteristics, see Figures 14 to 19.
[0184] Figures 14(a) and (b) are curves of vehicle density, speed and flow in human-machine mixed driving traffic flow under different penetration rates of connected autonomous vehicles. It can be seen from the figure that the speed continues to decrease with the increase of vehicle density. At the same time, with the increase of the penetration rate of connected autonomous vehicles, the average speed of road traffic flow under different vehicle densities increases. The mixing of connected autonomous vehicles helps to improve the overall operating speed of traffic flow; road flow gradually decreases after the vehicle density increases to the peak value. With the increase of the penetration rate of connected autonomous vehicles, the overall road flow gradually increases, and the vehicle density at the peak flow gradually increases with the increase of the penetration rate of connected autonomous vehicles. Therefore, the mixing of connected autonomous vehicles can greatly improve the road capacity.
[0185] Fig.15 The overall lane-changing frequency diagram of vehicles in mixed human-machine traffic flow is given by Fig.15It can be seen that the lane-changing frequency of vehicles in the traffic flow increases first and then decreases with the increase of the penetration rate of connected autonomous vehicles, and is mainly concentrated in the range of vehicle density 20~80veh / (km·ln) and connected autonomous vehicle penetration rate 0.3~0.8. This is mainly because the safety gap required for lane changing of connected autonomous vehicles is much smaller than that of manually driven vehicles. The mixing of connected autonomous vehicles increases the lane-changing frequency of vehicles to a certain extent. When the penetration rate P>0.8, the platoon driving characteristics of connected autonomous vehicles are obvious, the traffic flow operation state is relatively stable, and the lane-changing frequency of vehicles decreases significantly. Combined with Figure 14 (a) and (b), it can be seen that under the same density condition, with the increase of the penetration rate of connected autonomous vehicles, the average speed of traffic flow operation gradually increases, which leads to the continuous increase of the expected speed of vehicles, thus generating more frequent lane-changing intentions. At the same time, the lane-changing frequency of vehicles in the traffic flow also shows a trend of increasing first and then decreasing with the increase of vehicle density. Under different penetration rates of connected autonomous vehicles, when the vehicle density is on the left side of the white dotted line in the figure, the lane-changing frequency of vehicles gradually increases with the increase of vehicle density; as the vehicle density continues to increase, the lane-changing frequency of vehicles on the right side of the white dotted line gradually decreases. Combining Figure 14 (a) and (b), it can be seen that at this time, the average speed of traffic flow is low, and the road space occupancy rate is high, which makes it difficult for vehicles to change lanes, and the lane changing frequency gradually decreases.
[0186] Fig.16 The frequency diagram of coordinated lane changing of connected automatic vehicles in mixed human-machine driving traffic flow is shown in Figure 2. Fig.16 The coordinated lane-changing behavior of connected autonomous vehicles is mainly concentrated in the vehicle density range of 20 to 60 veh / (km·ln) and the penetration rate range of 0.2 to 0.8. When the penetration rate P is less than 0.3, the proportion of connected autonomous vehicles in the target lane is small, and it is difficult for connected autonomous vehicles to perform coordinated lane-changing behavior; when the penetration rate P is in the range of 0.2 to 0.8, connected autonomous vehicles appear more frequently in the target lane, and connected autonomous vehicles change lanes to the target lane through coordinated behavior; when the penetration rate P is greater than 0.8, most of the connected autonomous vehicles in the traffic flow are in a platoon driving state, and the frequency of coordinated lane-changing is significantly reduced. When the vehicle density is less than 40 veh / (km·ln), the frequency of coordinated lane-changing gradually increases with the increase of vehicle density, and reaches the maximum value at 40 veh / (km·ln); as the vehicle density continues to increase, the space conditions required for coordinated lane-changing are difficult to meet, and the frequency of coordinated lane-changing gradually decreases; when the vehicle density is greater than 100 veh / (km·ln), coordinated lane-changing behavior almost does not occur in the traffic flow.
[0187] Figures 17(a) to (d) are the spatiotemporal distribution trajectories of vehicles in the mixed human-machine driving traffic flow when the penetration rate of connected autonomous vehicles is 0, 0.3, 0.7 and 1 respectively; the traffic flow basically presents a stable operation state under the condition of 30veh / (km·ln). When the penetration rate P<0.3, the vehicle operation is relatively disordered, and there is a certain degree of congestion in the traffic flow; as the penetration rate gradually increases, when P=0.7, the vehicle trajectory gradually presents a regular operation, but there are still some vehicles in the traffic flow that stop and go, and their transmission with the simulation time has a certain degree of delay. When the penetration rate P=1, the traffic flow presents a stable operation, the congestion phenomenon is reduced and its transmission is extremely regular. This proves that connected autonomous vehicles can effectively stabilize the traffic flow and reduce the occurrence of vehicle stop and go.
[0188] Fig.18 is a curve chart of the penetration rate of connected autonomous driving vehicles and the average value of vehicle speed fluctuation at different vehicle speeds, Fig.18 It can be seen that when the vehicle density is less than 40veh / (km·ln), the average value of vehicle speed fluctuation gradually decreases with the increase of penetration rate. Networked autonomous driving vehicles can effectively reduce the discreteness of traffic flow speed and stabilize the traffic flow speed at the average level. As the vehicle density increases, the critical value of penetration rate that can play a significant stabilizing role in vehicle speed fluctuation gradually increases. This is because the increase in vehicle density reduces the average speed of traffic flow, and the absolute value of speed fluctuation decreases accordingly; and the traffic flow gradually enters a passive following state, and it is difficult to alleviate vehicle speed fluctuations with a low penetration rate. When the vehicle density is greater than 120veh / (km·ln), the fluctuation average value shows a trend of first increasing and then decreasing with the increase of penetration rate. Combined with the basic traffic flow diagram, it can be obtained that under this density condition, the purely manual driving traffic flow falls into a completely congested state. The increase in penetration rate increases the speed and flow of traffic flow, while leading to an increase in the average value of vehicle speed fluctuation.
[0189] Fig.19 The collision risk diagram of vehicles in mixed human-machine driving traffic flow is shown in the figure. The collision risk of vehicles in the driving process is represented by the ratio of the remaining time TTC of vehicles to be less than the threshold TTC*=3s. Fig.19 It can be obtained that under low vehicle density conditions, the traffic flow collision risk is low and the traffic flow is in a safe state; as the vehicle density increases, the vehicle collision risk gradually increases. When the penetration rate is low, the mixing of connected autonomous driving vehicles has almost no effect on improving the collision risk, but will cause fluctuations in the longitudinal collision risk. This phenomenon is due to the fact that the mixing of connected autonomous driving vehicles makes the traffic flow structure more complex, and the improvement effect of the lower penetration rate on the longitudinal collision risk of traffic flow is not enough to offset its own impact on traffic flow. When the penetration rate P ≥ 0.8, the formation driving and coordinated lane changing behaviors of connected autonomous driving vehicles become more frequent, the vehicle collision risk is significantly reduced, and the safety of traffic flow is significantly improved.
[0190] Any matters not described in the present invention are applicable to the prior art.
Claims
1. A method for constructing a lane-changing model for vehicles in mixed human-machine driving traffic flow, characterized in that: The lane-changing models for mixed human-machine traffic flows established by this method include a manually driven vehicle lane-changing model under a passive lane-changing intention scenario, a connected autonomous vehicle lane-changing model, and a connected autonomous vehicle lane-changing model under an active lane-changing intention scenario.
1. The lane-changing model of a manually driven vehicle in a passive lane-changing intention scenario is: 1) Based on the speed and position information of the vehicle ahead in the lane where the lane-changing vehicle is located, determine whether the vehicle has a passive lane-changing intention due to speed constraints. When the road conditions cannot meet the vehicle's expected driving speed at the next moment, the vehicle has a lane-changing intention. The speed constraint of the lane-changing vehicle is: in follow <v expect (1) In formula (1), v follow is the speed of the lane-changing vehicle constrained by the vehicle in front, v expect is the expected speed of the lane-changing vehicle at the next moment; 2) After the vehicle generates a passive lane-changing intention, it is determined whether the distance between the lane-changing vehicle and the target lane vehicle meets the safety distance according to formula (2); Gap n >Gap safe,n (2) In formula (2), Gap n Gap is the distance between the lane-changing vehicle and the target lane vehicle n. safe,n The safe distance required for lane-changing vehicles; The calculation formula for the safe distance between the lane-changing vehicle and the vehicle ahead in the target lane is: Among them, Gap safe,fd is the safe distance between the lane-changing vehicle and the preceding vehicle in the target lane; v m (t), v fd (t) are the speeds of the lane-changing vehicle m and the preceding vehicle fd in the target lane, respectively. fd is the braking deceleration of the front vehicle fd in the target lane, is the estimated vehicle deceleration, τ m is the lane-changing vehicle reaction time, d is the additional vehicle distance that the driver of a manually driven vehicle maintains when facing a connected autonomous vehicle. When the vehicle in the target lane is also manually driven, d = 0. Otherwise, it is calculated based on the driver's trust in the connected autonomous vehicle. The calculation formula for the safe distance between the lane-changing vehicle and the vehicle behind in the target lane is: Among them, Gap safe,rd is the safe distance between the lane-changing vehicle and the vehicle behind it in the target lane, v safe is the safe speed, b m is the braking deceleration of the lane-changing vehicle, x m (t), x rd (t) are the positions of the lane-changing vehicle and the vehicle behind in the target lane, v rd (t) is the speed of the vehicle behind in the target lane, τ rd is the reaction time of the vehicle behind in the target lane; 3) When a lane-changing vehicle has a passive lane-changing intention and the safety distance is met, the lane-changing vehicle should also consider the speed benefit condition. The speed benefit condition for the lane-changing vehicle is: v change >v n (t)(6) Among them, v change is the virtual speed of the vehicle after successfully changing lanes, v n (t) is the speed of vehicle n in the target lane; Therefore, after a manually driven vehicle generates a passive lane change intention, it can perform a free lane change if both the safety distance and speed benefit conditions are met; if not, it will continue to follow the vehicle.
2. The lane-changing model of a connected autonomous vehicle in a passive lane-changing intention scenario is: 1) After a connected autonomous vehicle generates a passive lane-changing intention, it can freely change lanes if the safety distance, lane-changing vehicle speed benefit conditions, and vehicle overall speed benefit conditions are simultaneously met. The speed constraint conditions, safety distance, and lane-changing vehicle speed benefit conditions for the connected autonomous vehicle to generate a passive lane-changing intention satisfy equations (1), (2), and (6), respectively. The vehicle overall speed benefit condition is as shown in equation (13): v change,rd >γv rd (t) (13) Among them, v change,rd is the virtual speed of the vehicle behind the target lane after the vehicle successfully changes lanes, v rd (t) is the speed of the vehicle behind the target lane; γ is the lane-changing decision parameter of the connected autonomous vehicle, which represents the threshold at which the lane-changing behavior of the connected autonomous vehicle reduces the speed of the vehicle behind the target lane, and its value range is 0–1; The calculation formula for the safe distance between the lane-changing vehicle and the vehicle ahead in the target lane is: The calculation formula for the safe distance between the lane-changing vehicle and the vehicle behind in the target lane is: 2) When a connected autonomous vehicle intends to passively change lanes, but road conditions do not meet the safety gap, the lane-changing vehicle exchanges information with other connected autonomous vehicles in the target lane to determine whether conditions for a coordinated lane change are met. If so, the vehicle coordinates the road conditions in the target lane through collaborative driving to meet the safety gap. If not, the lane-changing vehicle continues to follow in its lane. There are three scenarios for cooperative driving of vehicles in the target lane, namely: ① The front vehicle in the target lane is a connected autonomous driving vehicle or a connected autonomous driving vehicle queue, and the rear vehicle is a manually driven vehicle. The front vehicle in the target lane cooperates with the vehicle that changes lanes to complete the lane change; ② The front vehicle in the target lane is a manually driven vehicle, and the rear vehicle is a connected autonomous driving vehicle or a connected autonomous driving vehicle queue. The rear vehicle in the target lane cooperates with the vehicle that changes lanes to complete the lane change; ③ Both the front vehicle and the rear vehicle in the target lane are connected autonomous driving vehicles or a connected autonomous driving vehicle queue. The front and rear vehicles are regarded as connected autonomous driving vehicle queues respectively. When safety When the spacing satisfies both scenario ① and scenario ②, the connected autonomous driving vehicle platoon will disband and cooperate, and the vehicles ahead of the lane-changing vehicle in the longitudinal position will be accelerated as cooperative vehicles at the same time, and the cooperation process is the same as scenario ①; the vehicles behind the lane-changing vehicle in the longitudinal position will be decelerated as cooperative vehicles, and the cooperation process is the same as scenario ②; when the safety spacing satisfies scenario ① but does not satisfy scenario ②, the entire connected autonomous driving vehicle platoon will be accelerated as a cooperative vehicle, and the cooperation process is the same as scenario ①; when the safety spacing does not satisfy scenario ① but satisfies scenario ②, the entire connected autonomous driving vehicle platoon will be decelerated as a cooperative vehicle, and the cooperation process is the same as scenario ②; 3. The lane-changing model for a connected autonomous vehicle in the active lane-changing intention scenario is: An active lane-changing intention refers to the intention of a connected autonomous vehicle to change lanes when it is influenced by a queue of connected autonomous vehicles in the target lane. The lane-changing vehicle identifies the driving status of the vehicle in the current lane. If there is a queue of connected autonomous vehicles ahead and the queue has not reached the maximum queue length, the lane-changing vehicle maintains and follows the vehicle in the current lane. If there is a manually driven vehicle ahead or a full queue of connected autonomous vehicles, the lane-changing vehicle identifies the driving status of the vehicle in the target lane. If there is a manually driven vehicle ahead or a full queue of connected autonomous vehicles ahead in the target lane, the lane-changing vehicle maintains and follows the vehicle in the current lane. If there is a queue of connected autonomous vehicles ahead in the target lane and the queue has not reached the maximum queue length, the lane-changing vehicle position is divided into two cases: the lane-changing vehicle is located at the end or the middle of the queue of connected autonomous vehicles. When the lane-changing vehicle is at the tail end of a queue of connected autonomous vehicles, that is, the leading vehicle in the target lane is a connected autonomous vehicle and the following vehicle is a manually driven vehicle, the lane-changing process is the same as scenario ①. When the lane-changing vehicle is in the middle of a queue of connected autonomous vehicles, it is necessary to determine the distance between the lane-changing vehicle and the vehicle behind it in the queue. When the distance between the lane-changing vehicle and the vehicle behind it in the queue meets the safe distance, the lane-changing vehicle decelerates and drives to the tail end of the queue to change lanes. The safety distance that cooperative driving needs to meet in scenario ① is: Gap fd+1 >Gap safe,fd+1 (15) Gap fd+1 Gap is the distance between the front vehicle fd in the target lane and the vehicle fd+1 in the target lane. safe,fd+1 To meet the safety distance of the front vehicle fd in the target lane for cooperative driving; Gap fd+1 =x fd+1 (t)-x fd (t)-carL(16) x fd+1 (t), x fd (t) are the positions of the target lane vehicle fd+1 and the preceding vehicle fd, respectively. v fd+1 (t) is the speed of the vehicle fd+1 in the target lane, b fd+1 is the braking deceleration of vehicle fd+1 in the target lane, and carL is the vehicle length; When the target lane safety distance is met, the vehicle in front of the target lane adjusts its speed and drives in coordination with the lane-changing vehicle. The state of the vehicle in front of the target lane changes as shown in equations (18) to (20): v fd (t+Δt)=min[(v fd+1 (t)+a fd (t)Δt,v max ] (19) Among them, a fd (t) is the acceleration of the vehicle in front of the target lane, T is the expected time interval, v fd (t+Δt) is the speed of the vehicle in front of the target lane at the end of one simulation step of cooperative driving, Δt is the simulation step, v max is the maximum speed of the vehicle, x fd (t+Δt) is the position of the front vehicle fd in the target lane at the end of one simulation step cooperative driving, x' fd (t+Δt) is the position of the tail vehicle in the connected autonomous driving vehicle queue at the end of a simulation step cooperative driving, x' fd (t) is the position of the last vehicle in the connected autonomous vehicle queue at time t; After one simulation step, the safe distance Gap between the lane-changing vehicle and the preceding vehicle fd in the target lane is fd Update according to formula (21); Gap fd =x' fd (t+Δt)-(x m (t)+v m (t)Δt)(21) In scenario ②, the safety distance that cooperative driving needs to meet is: Gap rd-1 >Gap safe,rd-1 (22) Among them, Gap rd-1 Gap is the distance between the target lane vehicle rd and the target lane vehicle rd-1. safe,rd-1 To meet the safety distance of the target lane rear vehicle rd cooperative driving; Gap rd-1 =x' rd (t)-x rd-1 (t)-carL (23) x' rd (t) The position of the leading vehicle in the connected autonomous driving vehicle queue, x rd-1 (t) is the position of vehicle rd-1 in the target lane, v rd-1 (t) is the speed of the vehicle rd-1 in the target lane, τ rd-1 is the reaction time of vehicle rd-1 in the target lane, b rd-1 、b rd are the braking decelerations of the target lane vehicle rd-1 and the target lane vehicle rd, respectively; When the target lane safety distance is met, the vehicle behind the target lane adjusts its own speed and drives in coordination with the lane-changing vehicle. The state of the vehicle behind the target lane changes as shown in Equations (25) to (27): v rd (t+Δt)=max[(v rd (t)+a rd (t)Δt,0] (26) Among them, a rd (t) is the acceleration of the vehicle behind in the target lane, v rd (t+Δt), x rd (t+Δt) is the speed and position of the vehicle behind in the target lane at the end of a simulation step cooperative driving process, x' rd (t+Δt) is the position of the last vehicle in the connected autonomous driving vehicle queue at the end of a simulation step cooperative driving process; Gap m =x m (t)+v m (t)Δt-x rd (t+Δt) (28) After one simulation step, the safe distance Gap between the lane-changing vehicle and the vehicle behind in the target lane is m Update according to formula (28).
