An emergency vehicle route section priority passing control method considering a net-connected automatic driving special lane

Abstract

The application discloses an emergency vehicle road section priority passing control method considering a network-connected automatic driving special lane, and steps include: collecting vehicle state information of an emergency vehicle and all vehicles on a target road section; if the emergency vehicle enters the target road section, guiding the emergency vehicle to drive into the network-connected automatic driving special lane of the target road section, and calculating a dynamic emptying area in real time; regarding network-connected automatic driving vehicles in the dynamic emptying area as optimization vehicles, optimizing speed change values and lane changing decisions of the optimization vehicles by using a multi-objective optimization model established by taking minimization of emergency vehicle delay and minimization of interference on surrounding vehicles as targets, and then controlling the optimization vehicles by the optimized speed change values and lane changing decisions, so that network-connected automatic driving vehicles affecting passing in front of the emergency vehicle are emptied. The application optimizes longitudinal following and lateral lane changing decisions of network-connected automatic driving vehicles in social vehicles in a mixed traffic flow environment, ensures priority passing of the emergency vehicle, and reduces influence on the social vehicles.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of traffic control, in particular to an emergency vehicle route priority passing control method considering a net-connected automatic driving special lane. BACKGROUND

[0002] With the deepening of the aging degree and the deepening of the urbanization process, various types of sudden emergency rescue events occur frequently, and the travel delay of emergency vehicles is seriously increased. Although the law gives emergency vehicles higher road passing right, in the real urban road environment, this passive priority method will still be affected by surrounding vehicles, and the short-time lane-changing behavior to reduce the hindrance to emergency vehicles is more likely to cause traffic congestion, and it is difficult to achieve the priority passing of emergency vehicles. Therefore, it is of great significance to develop a scientific emergency vehicle priority passing control method to improve the driving efficiency of emergency vehicles while reducing the negative impact on surrounding traffic for the development of urban emergency rescue.

[0003] Currently, the priority passing control of emergency vehicles mainly focuses on path optimization, signal intersection priority passing control, and route priority passing control. Among them, for route priority passing control, existing strategies can be divided into the following three types: lane-changing strategy based on emergency vehicle sound and light alarm or vehicle networking technology; establishing temporary or permanent emergency vehicle special lane; strategy of using shoulder lane or reverse driving and other special ways. Compared with the research on path optimization and signal intersection priority passing control, the research on route priority passing control for emergency vehicles is less. However, on normal routes such as highway ramps or extended routes between intersections on urban roads, unlike signal intersections with clear timing schemes, they are more dependent on current road traffic flow conditions that are difficult to predict or estimate in advance, and emergency vehicles will be hindered by surrounding vehicles, resulting in deceleration and parking, and the uncertainty of travel time is large. Therefore, it is necessary to study the route priority passing strategy for emergency vehicles.

[0004] In recent years, with the rapid development of Internet of Vehicles technology, many scholars have proposed to divide the road right in space by setting a net-connected automatic driving special lane for the mixed traffic flow scenario that exists for a long time afterwards, and how to use the net-connected automatic driving special lane to control the route priority passing of emergency vehicles in the mixed traffic flow scenario is also a problem worth considering.

[0005] The Chinese patent application with the publication number CN116597674A proposes a dynamic emergency lane clearing method based on vehicle networking, which takes the upstream road section of the urban road intersection as the research object in the full-network environment, considers the queuing influence of the intersection, adopts a discretization method, and obtains the optimal clearing scheme and the optimal starting clearing time based on the principle of minimizing the clearing cost. Finally, the vehicle moves according to the clearing scheme to complete the emergency lane clearing. However, this scheme only considers the full-network environment and lacks research on mixed traffic flow environment. Moreover, this scheme adopts a discretization method, which greatly simplifies the complexity of the problem and is quite different from the real traffic flow environment. SUMMARY

[0006] The technical problem solved by the present application is to provide an emergency vehicle road section priority passing control method considering a networked automatic driving special lane, which reuses the networked automatic driving special lane as an emergency lane, optimizes the longitudinal following and lateral lane changing decisions of networked automatic driving vehicles, and ensures the passing efficiency of social vehicles under the premise of emergency vehicle priority passing.

[0007] To solve the above technical problems, the technical scheme adopted by the present application is:

[0008] An emergency vehicle road section priority passing control method considering a networked automatic driving special lane, comprising the following steps:

[0009] Collecting vehicle state information of the emergency vehicle and all vehicles on the target road section, and determining whether the emergency vehicle enters the target road section according to the vehicle state information of the emergency vehicle. If yes, guide the emergency vehicle to enter the networked automatic driving special lane of the target road section, and calculate the dynamic clearing area in front of the emergency vehicle in real time;

[0010] Establishing a multi-objective optimization model with the minimum delay of the emergency vehicle and the minimum interference to surrounding vehicles as the target, taking the networked automatic driving vehicles in the dynamic clearing area as optimization vehicles, and using the multi-objective optimization model to calculate the speed change value and lane changing decision of each optimization vehicle;

[0011] Using the speed change value and lane changing decision to control the optimization vehicles, thereby clearing the networked automatic driving vehicles in front of the emergency vehicle that affect the passing.

[0012] Further, when calculating the dynamic clearing area in front of the emergency vehicle in real time, it comprises:

[0013] Calculating the dynamic clearing distance in front of the emergency vehicle in real time. If the value of the dynamic clearing distance is less than or equal to a predetermined specified value, the value of the dynamic clearing distance is set to the specified value, and the expression of the dynamic clearing distance is as follows:

[0014]

[0015] wherein, D c is the dynamic clearance distance; e represents the emergency vehicle; v e (t) is the current speed of the emergency vehicle; T e is the time to ensure that the emergency vehicle continues to travel at the current speed; dec e is the maximum deceleration of the emergency vehicle; Gap min is the minimum stopping distance;

[0016] The rectangular region formed by the net-connected automatic driving special lane and the dynamic clearance distance in front of the emergency vehicle on the adjacent lane thereof is taken as the dynamic clearance region.

[0017] Further, the objective function expression of the multi-objective optimization model is as follows:

[0018] min O = w1*O1 + w2*O2 + w3*(O3 + O4)

[0019] wherein, w1, w2, and w3 are the weights of each part in the objective function, respectively;

[0020] O1 is the distance of the emergency vehicle from the exit of the road section, and the expression is as follows:

[0021]

[0022] wherein, L is the length of the road section; t is the current time; T0 is the optimization step; x e (t) is the position of the emergency vehicle at t; v e (t) is the speed of the emergency vehicle at t; is the optimized speed change value of the emergency vehicle at t;

[0023] O2 is the number of optimized vehicles among all optimized vehicles that are still on the net-connected automatic driving special lane and have not driven off the target road section after this optimization, and the expression is as follows:

[0024]

[0025] wherein, ω represents the optimized vehicle; k is the lane index; represents whether the optimized vehicle is still on lane k after optimization at t, and is 1 if it is on lane k, otherwise 0; p ω (t+T0) is a 0-1 variable;

[0026] O3 is the negative sum of the speed change values of all optimized vehicles after this optimization, and the expression is as follows:

[0027]

[0028] wherein, is the speed change value of the optimization vehicle in this optimization;c ω (t+T0) is an integer variable;

[0029] O4 is the opposite number of the sum of all positive vehicle speed change values of the optimization vehicle after this optimization, expressed as follows:

[0030]

[0031] wherein, is the speed change value of the emergency vehicle in this optimization;c ω (t+T0) is an integer variable.

[0032] Further, the constraint conditions of the multi-objective optimization model include vehicle dynamics constraints, expressed as follows:

[0033]

[0034] 0≤v ω (t+T0)≤vmax ω

[0035] dec ω ≤a ω (t+T0)≤acc ω

[0036] wherein, ω represents the optimization vehicle; t is the current time; T0 is the optimization step; x ω (t+T0) is the position of the optimization vehicle after optimization at t; x ω (t) is the position of the optimization vehicle at t; v ω (t) is the speed of the optimization vehicle at t; is the speed change value of the optimization vehicle at t; v ω (t+T0) is the speed of the optimization vehicle after optimization at t; a ω (t+T0) is the acceleration of the optimization vehicle after optimization at t; vmax ω is the maximum speed of the optimization vehicle; dec ω is the maximum deceleration of the optimization vehicle; acc ω is the maximum acceleration of the optimization vehicle.

[0037] Further, the constraint conditions of the multi-objective optimization model include lane constraints, expressed as follows:

[0038]

[0039] wherein, ω represents the optimization vehicle; k is the lane index; represents whether the optimization vehicle is still on lane k after optimization at t, if on lane k it is 1, otherwise 0.

[0040] Further, the constraint condition of the multi-objective optimization model includes a following safety distance constraint, expressed as follows:

[0041]

[0042] d1 ω (t+T0)≥max{ds1 ω (t+T0),0}

[0043]

[0044] d2 ω (t+T0)≥max{ds2 ω (t+T0),0}

[0045] where ω represents an optimization vehicle; ds1 ω (t+T0) is a safety distance between the optimization vehicle and a front vehicle in the same lane of the optimization vehicle after optimization at t; TTC min is a collision time threshold limit as a substitute safety index; v ω (t+T0) is a speed of the optimization vehicle after optimization at t; is a speed of the front vehicle in the same lane of the optimization vehicle after optimization at t; d1 ω (t+T0) is an actual distance between the optimization vehicle and the front vehicle in the same lane of the optimization vehicle after optimization at t; is a position of the front vehicle in the same lane of the optimization vehicle after optimization at t; x ω (t+T0) is a position of the optimization vehicle after optimization at t; is a vehicle length of the front vehicle in the same lane of the optimization vehicle; ds2 ω (t+T0) is a safety distance between the optimization vehicle and a rear vehicle in the same lane of the optimization vehicle after optimization at t; is a speed of the rear vehicle in the same lane of the optimization vehicle after optimization at t; d2 ω (t+T0) is an actual distance between the optimization vehicle and the rear vehicle in the same lane of the optimization vehicle after optimization at t; is a position of the rear vehicle in the same lane of the optimization vehicle after optimization at t; l ω is a vehicle length of the optimization vehicle.

[0046] Further, the constraint condition of the multi-objective optimization model includes a lane changing safety distance constraint, expressed as follows:

[0047]

[0048] d3 ω (t+T0)≥max{ds3 ω (t+T0),0}

[0049]

[0050] d4 ω (t+T0)≥max{ds4 ω (t+T0),0}

[0051] where ω represents the optimization vehicle; ds3 ω (t+T0) is the safety distance between the optimization vehicle and the front vehicle in the target lane after optimization at t; TTC min is the collision time threshold limit as an alternative safety index; v ω (t+T0) is the speed of the optimization vehicle after optimization at t; is the speed of the front vehicle in the target lane of the optimization vehicle after optimization at t; d3 ω (t+T0) is the actual distance between the optimization vehicle and the front vehicle in the target lane after optimization at t; is the position of the front vehicle in the target lane of the optimization vehicle after optimization at t; x ω (t+T0) is the position of the optimization vehicle after optimization at t; is the vehicle length of the front vehicle in the target lane of the optimization vehicle; ds4 ω (t+T0) is the safety distance between the optimization vehicle and the rear vehicle in the target lane after optimization at t; is the speed of the rear vehicle in the target lane of the optimization vehicle after optimization at t; d4 ω (t+T0) is the actual distance between the optimization vehicle and the rear vehicle in the target lane after optimization at t; is the position of the rear vehicle in the target lane of the optimization vehicle after optimization at t; l ω is the vehicle length of the optimization vehicle.

[0052] Further, the constraint conditions of the multi-objective optimization model include auxiliary variable constraints, expressed as follows:

[0053]

[0054] where ω represents the optimization vehicle; v ω (t+T0) is the speed of the optimization vehicle after optimization at t; x ω (t+T0) is the position of the optimization vehicle after optimization at t; auxiliary variable c ω (t+T0) is an integer variable, -1 represents that the optimization vehicle accelerates after this optimization; 0 represents that the optimization vehicle maintains the speed after this optimization; 1 represents that the optimization vehicle decelerates after this optimization; auxiliary variable ρ ω (t+T0) is a binary variable, 1 represents that the optimization vehicle does not leave the target section after this optimization; 0 represents that the optimization vehicle leaves the target section.

[0055] Further, before using the multi-objective optimization model to calculate the speed change value and lane change decision of each optimization vehicle, comprising: configuring a corresponding car-following model and lane change model for all vehicles on the target road section; when using the multi-objective optimization model to calculate the speed change value and lane change decision of each optimization vehicle, comprising:

[0056] The artificial driving vehicles on the target road section and the connected automatic driving vehicles outside the dynamic emptying area are regarded as non-optimization vehicles, the vehicle state information of the non-optimization vehicles at the current time step is input into the corresponding car-following model and lane change model respectively, and the vehicle state information of the non-optimization vehicles after one optimization step is calculated;

[0057] The vehicle state information of the optimization vehicles at the current time step and the vehicle state information of the non-optimization vehicles after one optimization step are input into the multi-objective optimization model and solved, and the speed change value and lane change decision of each optimization vehicle are obtained.

[0058] Further, when using the speed change value and the optimized lane change decision to control each optimization vehicle respectively, comprising:

[0059] If the optimization step of the multi-objective optimization model is greater than the control step, the speed change value is decomposed into each control step to control the speed change of the corresponding optimization vehicle through linear interpolation, and the lane change decision is controlled in the last control step of the corresponding optimization step to control the position change of the corresponding optimization vehicle.

[0060] Compared with the prior art, the application has the advantages that:

[0061] The application aims at a mixed traffic flow environment, temporarily opens a connected automatic driving exclusive lane as an emergency lane, dynamically divides the road right from space by reusing the exclusive lane as a dynamic emergency lane. When an emergency vehicle enters the road section, it can pass through the connected automatic driving exclusive lane first, and the dynamic emptying distance in front of the emergency vehicle is calculated, and the longitudinal car-following and lateral lane change decisions of the connected vehicles within the dynamic emptying distance are optimized to realize the avoidance strategy based on cooperative lane changing, timely empty the area in front of the emergency vehicle, and realize the priority of the emergency vehicle.

[0062] The application takes the minimum delay of the emergency vehicle and the least interference to the surrounding vehicles as the target, takes the longitudinal speed and lane change behavior of the connected vehicle as the decision variable, considers the vehicle dynamics, car-following and lane change safety distance constraints, establishes a multi-objective optimization model to obtain the speed of all connected automatic driving vehicles CAV and the lane change decision of the connected automatic driving vehicles CAV in front of the emergency vehicle on the exclusive lane, and decomposes the optimization result to each control step for control. While ensuring the priority of the emergency vehicle, the influence on the social vehicles is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0063] Figure 1Flowchart for the embodiment of the present application.

[0064] Figure 2 Schematic diagram for a road section scenario.

[0065] Figure 3 Statistical diagram for average speed of emergency vehicles.

[0066] Figure 4 Statistical diagram for number of times of CAV lane changing for assisting emergency vehicle priority. DETAILED DESCRIPTION

[0067] The present application is further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the protection scope of the present application is not limited thereby.

[0068] The present embodiment proposes an emergency vehicle road section priority passing control method considering a networked automatic driving exclusive lane, and realizes priority passing of an emergency vehicle on a target road section in a mixed traffic flow environment. The overall idea is to reuse the networked automatic driving exclusive lane as an emergency lane when the emergency vehicle enters the target road section, and to pre-empty the vehicles within the dynamic emptying distance in front of the emergency vehicle by optimizing their car following and lane changing decisions, thereby ensuring the priority passing of the emergency vehicle while reducing the impact on social vehicles.

[0069] As shown in Figure 1 , the method of the present embodiment includes the following steps:

[0070] S1) Collect vehicle state information of all social vehicles and emergency vehicles on the target road section. In the present embodiment, the social vehicles include human-driven vehicles (HDV) and networked automatic driving vehicles (CAV), and the vehicle state information includes speed, acceleration, position and other information of the vehicles;

[0071] According to the position information in the vehicle state information of the emergency vehicle, it is determined whether the emergency vehicle enters the target road section. If the emergency vehicle enters the target road section, the networked automatic driving exclusive lane of the target road section is reused as an emergency lane, and the emergency vehicle is guided to enter the emergency lane. At the same time, by calculating the dynamic emptying distance in front of the emergency vehicle, the dynamic emptying area of this time, also known as the optimization area, is determined;

[0072] S2) A multi-objective optimization model is established with the minimum delay of the emergency vehicle and the minimum disturbance to the surrounding vehicles as the target, considering the constraints of vehicle dynamics, car following and lane changing safety distance, etc. The networked automatic driving vehicles in the dynamic emptying area are taken as optimization vehicles, and the speed change value and lane changing decision of the optimization vehicles are calculated as the optimization result corresponding to each optimization vehicle using the multi-objective optimization model;

[0073] S3) According to the optimization result, the optimization vehicle control is performed, so as to empty the networked automatic driving vehicles in front of the emergency vehicle that affect the passing.

[0074] Through the above steps, when the emergency vehicle enters the road section, first, the dedicated lane for networked automatic driving is reused as an emergency lane, the dynamic emptying distance of the emergency vehicle at the current time is calculated, the networked automatic driving vehicle located in the optimization area that needs to be optimized is determined, second, a multi-objective optimization model is established with the minimum delay of the emergency vehicle and the minimum interference to surrounding vehicles as the target, considering the constraints of vehicle dynamics, following and lane changing safety distance, etc., the speed change value and lane changing decision of the optimized vehicle are calculated, and finally, the control is performed according to the optimization result, and the optimization result is decomposed to each control step and reverse control. Only by controlling the following and lane changing decision of the vehicle within a certain dynamic distance in front of the emergency vehicle, the networked automatic driving vehicle affecting the passing of the emergency vehicle is pre-cleared, and the delay of the emergency vehicle is reduced. It is of great help to the problem of emergency vehicle road section priority passing under mixed traffic flow, and also realizes the minimum interference to the social vehicle.

[0075] The following will be described in detail.

[0076] In step S1 of the embodiment, all vehicles on the target road section are configured with corresponding following models and lane changing models, after the vehicle state information is collected, the following steps are performed:

[0077] S11) According to the position information, it is judged whether the emergency vehicle enters the road section, if the emergency vehicle has not entered, all vehicles drive according to their following models and lane changing models; if the emergency vehicle has entered, the networked automatic driving lane is reused as an emergency lane and the emergency vehicle is guided to enter the emergency lane;

[0078] S12) The dynamic emptying distance of the emergency vehicle at the current time is calculated, in the embodiment, three parts are considered when calculating the dynamic emptying distance: the distance required for the emergency vehicle to drive at the current speed for T e seconds, the distance required for the emergency vehicle to decelerate from the current speed to 0, and the minimum parking distance. The expression of the dynamic emptying distance is as follows:

[0079]

[0080] Where D2 is the calculated value of the dynamic emptying distance of the emergency vehicle; the emergency vehicle is defined as e, and the optimized vehicle is defined as ω; v e (t) is the current speed of the emergency vehicle; T e is the time to ensure that the emergency vehicle continues to drive at the current speed, which is set to 2s in the embodiment; dec e is the maximum deceleration of the emergency vehicle; Gap min is the minimum parking distance, which is set to 2m in the embodiment;

[0081] If the emptying distance length is too short, the vehicle may not be emptied in time, and the priority of the emergency vehicle cannot be guaranteed. In this case, when the dynamic emptying distance D c of the emergency vehicle is too short, the specified 50 meters is taken as the value of the dynamic emptying distance of the emergency vehicle, and the expression is as follows:

[0082]

[0083] where D e is the final value of the dynamic emptying distance of the emergency vehicle;

[0084] S13) According to the calculated dynamic emptying distance of the emergency vehicle, the dynamic emptying area is determined as a rectangular area formed by the dynamic emptying distance in front of the emergency vehicle on the connected automatic driving exclusive lane and its adjacent lanes, and the connected automatic driving vehicles in the area are the optimized vehicles for the next step.

[0085] Step S2 of the embodiment uses a multi-objective optimization model to solve the longitudinal following and lateral lane changing decisions of the optimized vehicles according to the collected data, achieving the goal of minimizing the delay of the emergency vehicle and minimizing the interference to the surrounding vehicles, including the following steps:

[0086] S21) Taking minimizing the delay of the emergency vehicle and minimizing the interference to the surrounding vehicles as the goal, considering the vehicle dynamics, following and lane changing safety distance constraints, a multi-objective optimization model is established to calculate the speed change value and lane changing decision of the optimized vehicle;

[0087] S22) The manually driven vehicles on the target section and the connected automatic driving vehicles outside the dynamic emptying area are used as non-optimized vehicles, and the non-optimized vehicles are used as the front and rear vehicles of the optimized vehicles in the multi-objective optimization model. According to the vehicle position, speed and acceleration information collected at the current time step, the vehicle state of the non-optimized vehicle after one optimization step is obtained through the corresponding following and lane changing model of the vehicle.

[0088] S23) For each optimized vehicle, the vehicle state information of the optimized vehicle at the current time step and the vehicle state information of the corresponding non-optimized vehicle after one optimization step are input into the multi-objective optimization model and solved to obtain the optimal solution of the speed change value and lane changing decision of each optimized vehicle.

[0089] In this embodiment, the objective function of the multi-objective optimization model contains four parts; the constraint conditions of the multi-objective optimization model include vehicle dynamics constraints, lane constraints, following safety distance constraints, lane changing safety distance constraints, and auxiliary variable constraints, which are described in detail below.

[0090] The objective function of the multi-objective optimization model consists of four parts, and the expression is as follows:

[0091] minO = w1*O1 + w2*O2 + w3*(O3 + O4)

[0092] wherein w1, w2, w3 are the weights of each part of the objective function, and in the embodiment, w1 = 400, w2 = 200, w3 = 1.

[0093] The first part O1 of the objective function is the distance of the emergency vehicle from the exit of the road section, and the expression is as follows:

[0094]

[0095] wherein L is the length of the road section; t is the current time; T0 is the optimization step; x e (t) is the position of the emergency vehicle at t; v e (t) is the speed of the emergency vehicle at t; is the optimized speed change value of the emergency vehicle at t.

[0096] The second part O2 of the objective function is the number of optimized vehicles among all optimized vehicles that are still on the dedicated lane for networked automatic driving after this optimization and have not yet driven off the target road section, and the expression is as follows:

[0097]

[0098] wherein ω represents the optimized vehicle; k is the lane index, respectively 0, 1, 2; k = 0 represents being on lane 0, i.e. on the dedicated lane for networked automatic driving; represents whether the optimized vehicle is still on lane k after optimization at t, if on lane k, it is 1, otherwise 0; p ω (t+T0) is a 0-1 variable.

[0099] The third part O3 of the objective function is the inverse of the sum of the negative speed change values of all optimized vehicles after this optimization, and the expression is as follows:

[0100]

[0101] wherein is the speed change value of the optimized vehicle in this optimization; c ω (t+T0) is an integer variable, taking values -1, 0, 1.

[0102] The fourth part O4 of the objective function is the inverse of the sum of the positive speed change values of all optimized vehicles after this optimization, and the expression is as follows:

[0103]

[0104] wherein is the speed change value of the optimized vehicle in this optimization; c ω(t+T0) is an integer variable, taking values -1, 0, 1.

[0105] In the constraint condition of the multi-objective optimization model, the vehicle dynamics constraint of the embodiment ensures that the generated control strategy is physically reasonable, the system is stable, the driving is safe, and the overall performance of the vehicle can be optimized, and the expression is as follows:

[0106]

[0107] 0≤v ω (t+T0)≤vmax ω

[0108] dec ω ≤a ω (t+T0)≤acc ω

[0109] Wherein, ω represents an optimized vehicle; t is a current time; T0 is an optimization step; x ω (t+T0) is a position of the optimized vehicle after optimization at t; x ω (t) is a position of the optimized vehicle at t; ω (t) is a speed of the optimized vehicle at t; is a speed change value of the optimized vehicle at t; v ω (t+T0) is a speed of the optimized vehicle after optimization at t; a ω (t+T0) is an acceleration of the optimized vehicle after optimization at t; vmax ω is a maximum speed of the optimized vehicle; dec ω is a maximum deceleration of the optimized vehicle; acc ω is a maximum acceleration of the optimized vehicle.

[0110] In the constraint condition of the multi-objective optimization model, the lane constraint of the embodiment ensures that each vehicle only appears in one lane at each time, and when the optimized vehicle ω is located in the k lane at t, the optimized vehicle ω only changes lanes to the left and ensures that there is no continuous lane change, and the expression is as follows:

[0111]

[0112] Wherein, ω represents an optimized vehicle; k is a lane index; represents whether the optimized vehicle is still in the lane k after optimization at t, if it is in the lane k, it is 1, otherwise it is 0.

[0113] In the constraint condition of the multi-objective optimization model, the following safety distance constraint of the embodiment uses the threshold limit of the replacement safety index collision time (TTC) between the vehicle and the front and rear vehicles in the lane, and the expression is as follows:

[0114]

[0115] d1 ω (t+T0)≥max{ds1 ω (t+T0),0}

[0116]

[0117] d2 ω (t+T0)≥max{ds2 ω (t+T0),0}

[0118] where ω represents the optimization vehicle; ds1 ω (t+T0) is the safety distance between the optimization vehicle and the front vehicle of the lane of the optimization vehicle after optimization at t; TTC min is the collision time threshold limit as the alternative safety index, which is set to 2s in this embodiment; v ω (t+T0) is the speed of the optimization vehicle after optimization at t; v ω (t+T0) is the speed of the optimization vehicle after optimization at t; is the speed of the front vehicle of the lane of the optimization vehicle after optimization at t; d1 ω (t+T0) is the actual distance between the optimization vehicle and the front vehicle of the lane of the optimization vehicle after optimization at t; is the position of the front vehicle of the lane of the optimization vehicle after optimization at t; x ω (t+T0) is the position of the optimization vehicle after optimization at t; is the length of the front vehicle of the lane of the optimization vehicle; ds2 ω (t+T0) is the safety distance between the optimization vehicle and the rear vehicle of the lane of the optimization vehicle after optimization at t; is the speed of the rear vehicle of the lane of the optimization vehicle after optimization at t; d2 ω (t+T0) is the actual distance between the optimization vehicle and the rear vehicle of the lane of the optimization vehicle after optimization at t; is the position of the rear vehicle of the lane of the optimization vehicle after optimization at t; l ω is the length of the optimization vehicle.

[0119] In the constraint condition of the multi-objective optimization model, the lane-changing safety distance constraint of this embodiment uses the need to meet the collision time (Time-to-Collision, TTC) threshold limit between the ego vehicle and the front and rear vehicles of the target lane, and the expression is as follows:

[0120]

[0121] d3 ω (t+T0)≥max{ds3ω (t+T0),0}

[0122]

[0123] d4 ω (t+T0)≥max{ds4 ω (t+T0),0}

[0124] where ω represents the optimization vehicle; ds3 ω (t+T0) is the safety distance between the optimization vehicle and the front vehicle of the target lane after optimization at t; TTC min is the collision time threshold limit as an alternative safety index, which is set to 2s in this embodiment; v ω (t+T0) is the speed of the optimization vehicle after optimization at t; is the speed of the front vehicle of the target lane of the optimization vehicle after optimization at t; d3 ω (t+T0) is the actual distance between the optimization vehicle and the front vehicle of the target lane after optimization at t; is the position of the front vehicle of the target lane of the optimization vehicle after optimization at t; x ω (t+T0) is the position of the optimization vehicle after optimization at t; is the vehicle length of the front vehicle of the target lane of the optimization vehicle; ds4 ω (t+T0) is the safety distance between the optimization vehicle and the rear vehicle of the target lane after optimization at t; is the speed of the rear vehicle of the target lane of the optimization vehicle after optimization at t; d4 ω (t+T0) is the actual distance between the optimization vehicle and the rear vehicle of the target lane after optimization at t; is the position of the rear vehicle of the target lane of the optimization vehicle after optimization at t; l ω is the vehicle length of the optimization vehicle.

[0125] In the constraint conditions of the multi-objective optimization model, the auxiliary variable constraints of this embodiment can improve the accuracy and control effect of the model, and also ensure physical feasibility, simplify the solution of the optimization problem, and the expression is as follows:

[0126]

[0127] where ω represents the optimization vehicle; auxiliary variable c ω (t+T0) is an integer variable, -1 represents that the optimization vehicle accelerates after this optimization; 0 represents that the optimization vehicle does not change speed after this optimization; and 1 represents that the optimization vehicle decelerates after this optimization.

[0128]

[0129] where auxiliary variable pω (t+T0) is a binary variable, 0 represents that the optimized vehicle drives off the road section after this optimization; 1 represents that the optimized vehicle does not drive off the road section.

[0130] From the above, the non-optimized vehicle can be regarded as an obstacle in the optimization process of the optimized vehicle, and is mainly used as the front and rear vehicles of the optimized vehicle in the car-following safety distance constraint and the lane-changing safety distance constraint in the optimization model, for example: and the like. At the same time, because the optimized vehicle needs to satisfy the safety distance with the surrounding vehicles in the next optimization step in the optimization process, the vehicle state of the non-optimized vehicle after one optimization step is taken as the input of the multi-objective optimization model.

[0131] In step S3 of the embodiment, the optimized vehicle control is performed according to the optimization result. Specifically, when the model optimization step is greater than the control step, the speed change value of the optimized vehicle is decomposed to each control step by linear interpolation for control, and the lane-changing decision of the optimized vehicle is controlled in the last control step within the optimization step. In the embodiment, the control step is set to 0.1 s. Since the model optimization step is 1 s and the control step is 0.1 s, the specific steps include the following steps:

[0132] S31) Linear interpolation is performed on the solved speed change value of the optimized vehicle, and the speed change value corresponding to each control step is used for speed control of the optimized vehicle, so as to realize speed adjustment of the optimized vehicle;

[0133] S32) The lane-changing decision of the optimized vehicle solved is controlled in the last control step within the corresponding optimization step, so as to realize position adjustment of the optimized vehicle.

[0134] The effect of the method of the embodiment is verified through specific tests.

[0135] A 1.2-kilometer three-lane urban expressway section is selected, as shown in Figure 2 .

[0136] In the embodiment, SUMO and PYTHON are combined for simulation, and vehicle parameters of different vehicle types are set, as shown in Table 1.

[0137] Table 1 Vehicle parameter settings of different vehicle types

[0138]

[0139] In the embodiment, two schemes are set for comparison to verify the effectiveness of the proposed model. Scheme one is the basic scheme; scheme two is the scheme proposed in the present application. The scheme setting comparison is shown in Table 2.

[0140] Table 2 Scheme setting comparison

[0141]

[0142] At the same time, different traffic, permeability, emergency vehicle entering time and other scenarios are simulated and evaluated in this embodiment. The parameter settings are as follows: the traffic value range is 900veh·h -1 -1500veh·h -1 , interval 200veh·h -1 ; the permeability value range is 40%-80%, interval 10%; the emergency vehicle entering time value range is 80s-170s, interval 10s. The simulation experiment of two schemes is 200(4*5*10) groups.

[0143] By executing the method of this embodiment, the current step vehicle state of the optimized vehicle and the non-optimized vehicle is obtained from SUMO, and then the vehicle state of the non-optimized vehicle after one optimization step is calculated according to the current step vehicle state of the non-optimized vehicle using the corresponding car following and lane changing model. The vehicle state of the non-optimized vehicle after one optimization step and the current step vehicle state of the optimized vehicle are used as the input of the optimization model for subsequent solving and optimization vehicle control.

[0144] Compared with the scheme without optimization model, the method proposed in this embodiment can reduce the average blocking time of the emergency vehicle from 63.35 seconds to 2.50 seconds, as shown in Table Three; the average travel delay of the emergency vehicle is reduced from 12.46 seconds to 0.87 seconds, as shown in Table Four; since the maximum speeds of the CAV vehicle and the emergency vehicle are 16m / s and 20m / s respectively, the average speed of the emergency vehicle in scheme one is almost 16m / s, which means that the emergency vehicle is mostly blocked by the front CAV vehicle. The average speed of the emergency vehicle in scheme two can almost reach 20m / s, which means that the emergency vehicle is basically not blocked by the CAV vehicle and can maintain the maximum speed throughout the journey, as shown in Table Five; the proposed scheme can optimize the lane changing of 6 CAV vehicles on average for the emergency vehicle under the above simulation scenarios, and the effectiveness of the model application is verified, as shown in Table Six; and the average speed of the vehicle flow except the emergency vehicle changes little, which means that the implementation of the scheme does not significantly reduce the traffic efficiency of the surrounding vehicles and has little effect on the surrounding vehicles, as shown in Table Five. Figure 3 Figure 4

[0145] Table Three Average blocking time of emergency vehicle (s)

[0146]

[0147] Table Four Average travel delay of emergency vehicle (s)

[0148] ​​

[0149] Table five average speed of vehicle flow (m / s) except emergency vehicles -1 )

[0150]

[0151]

[0152] In summary, the present application considers the emergency vehicle priority problem of road section, and proposes an emergency vehicle road section priority control method considering the network-connected automatic driving special lane, which multiplexes the network-connected automatic driving special lane as a dynamic emergency lane, and realizes the dynamic division of road right in space. Secondly, taking the minimum delay of emergency vehicles and the minimum interference of surrounding vehicles as the target, considering the vehicle dynamics, following and lane changing safety distance constraints, a multi-objective optimization model is established to calculate the speed change value and lane changing decision of the optimized vehicle. Finally, according to the optimization result, the speed change value of the optimized vehicle is decomposed to each control step through linear interpolation for control, and the lane changing decision of the optimized vehicle is controlled in the last control step within the optimization step. While ensuring the priority of emergency vehicles, the influence on social vehicles is reduced, and a new optimization idea is provided for the research of emergency vehicle road section priority control method in the mixed driving scene which will appear for a long time.

[0153] The above only describes the preferred embodiments of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solution falling within the concept of the present application shall fall within the protection scope of the present application. It should be noted that for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application shall also be considered as the protection scope of the present application.

Claims

1.A method for controlling the priority of an emergency vehicle on a road section considering a dedicated lane for connected and automated vehicles, characterized in that, The method comprises the following steps: Collecting vehicle state information of the emergency vehicle and all vehicles on the target road section, and determining whether the emergency vehicle enters the target road section according to the vehicle state information of the emergency vehicle, and if so, guiding the emergency vehicle to enter a network-connected automatic driving exclusive lane of the target road section, and calculating a dynamic emptying area in front of the emergency vehicle in real time; A multi-objective optimization model is established with the minimum delay of the emergency vehicle and the minimum interference to surrounding vehicles as the target, and the objective function expression of the multi-objective optimization model is as follows: wherein, , , are the weights of each part in the objective function, respectively. The distance of the emergency vehicle from the exit of the road segment is expressed as follows: wherein, is the length of the road segment; is the current time; is the optimization step size, set to 1 second; is the position of the emergency vehicle at ; is the speed of the emergency vehicle at ; is the speed change value optimized for the emergency vehicle at ; For all the optimized vehicles which are still on the connected and automated driving special lane and have not left the target section after this optimization, the expression is as follows: wherein, denotes an optimized vehicle; is a lane index; denotes whether the optimized vehicle is still on lane k after optimization, 1 if on lane k, otherwise 0; denotes whether the optimized vehicle is still on lane k after optimization, 1 if on lane k, otherwise 0; is a 0-1 variable; To all the optimized vehicles, the sum of the negative vehicle speed change values after this optimization is expressed as follows: wherein, is an integer variable; and is an integer variable; and To optimize the sum of the positive vehicle speed change values of all vehicles after this optimization, the opposite number is expressed as follows: wherein, is the speed change value for the emergency vehicle in this optimization; is an integer variable; The network-connected automatic driving vehicles in the dynamic emptying area are used as optimization vehicles, and the multi-objective optimization model is used to calculate the speed change value and lane changing decision of each optimization vehicle; The speed change value and the optimized lane changing decision are used to control each optimization vehicle, so as to empty the network-connected automatic driving vehicles in front of the emergency vehicle that affect the traffic. 2.The method of claim 1, wherein, When calculating the dynamic emptying area in front of the emergency vehicle in real time, the following steps are included: The dynamic emptying distance in front of the emergency vehicle is calculated in real time, and if the value of the dynamic emptying distance is less than or equal to a preset specified value, the value of the dynamic emptying distance is set to the specified value, and the expression of the dynamic emptying distance is as follows: wherein, is the dynamic clearance distance; represents an emergency vehicle; is the speed of the emergency vehicle at ; is the time to guarantee the emergency vehicle to continue to travel at the current speed ; is the maximum deceleration of the emergency vehicle; is the minimum stopping distance; A rectangular area formed by the dynamic emptying distance in front of the emergency vehicle on the network-connected automatic driving exclusive lane and its adjacent lanes is used as the dynamic emptying area. 3.The method of claim 1, wherein, The constraint conditions of the multi-objective optimization model include vehicle dynamics constraints, and the expression is as follows: in, Indicates optimized vehicle; For the current moment; To optimize the step size, set it to 1s; To optimize the vehicle The optimized position; To optimize the vehicle The position at the time; To optimize the vehicle speed at 1 hour; To optimize the vehicle Optimized speed change value; To optimize the vehicle Optimized speed; To optimize the vehicle Optimized acceleration; To optimize the maximum speed of the vehicle; To optimize the maximum deceleration of the vehicle; To optimize the vehicle's maximum acceleration. 4.The method of claim 1, wherein, The constraint conditions of the multi-objective optimization model include lane constraints, and the expression is as follows: in, Indicates optimized vehicle; Index for lanes; Indicates that the optimized vehicle Is it still on lane k after optimization? If it is on lane k, it is 1, otherwise it is 0. 5.The method of claim 1, wherein, The constraint conditions of the multi-objective optimization model include following safety distance constraints, and the expression is as follows: in, Indicates optimized vehicle; To optimize the vehicle The safe distance between the vehicle in front of the lane after optimization; As an alternative safety indicator, the collision time threshold limit is used; To optimize the vehicle Optimized speed; To optimize the vehicle's lane, the vehicle ahead is in Optimized speed; To optimize the vehicle The actual distance between the vehicle in front of the lane after optimization; To optimize the vehicle's lane, the vehicle ahead is in The optimized position; To optimize the vehicle The optimized position; To optimize the length of the vehicle in front of the vehicle in the lane; To optimize the vehicle The safe distance between the vehicle behind the lane after optimization; To optimize the vehicle's lane, the following vehicle is in Optimized speed; To optimize the vehicle The actual distance between the vehicle and the following vehicle in the lane after optimization; To optimize the vehicle's lane, the following vehicle is in The optimized position; To optimize the vehicle's length. 6.The method of claim 1, wherein, The constraint conditions of the multi-objective optimization model include lane changing safety distance constraints, and the expression is as follows: in, Indicates optimized vehicle; To optimize the vehicle The safe distance between the vehicle and the preceding vehicle in the target lane after optimization; As an alternative safety indicator, the collision time threshold limit is used; To optimize the vehicle Optimized speed; To optimize the target lane of the vehicle, the vehicle ahead is in Optimized speed; To optimize the vehicle The actual distance between the vehicle in front and the target lane after optimization; To optimize the target lane of the vehicle, the vehicle ahead is in The optimized position; To optimize the vehicle The optimized position; To optimize the length of the vehicle ahead in the target lane; To optimize the vehicle The safe distance between the vehicle and the target lane after optimization; To optimize the target lane of the vehicle, the following vehicle is in Optimized speed; To optimize the vehicle The actual distance between the vehicle and the target lane after optimization; To optimize the vehicle's lane, the following vehicle is in The optimized position; To optimize the vehicle's length. 7.The method of claim 1, wherein, The constraint conditions of the multi-objective optimization model include auxiliary variable constraints, and the expression is as follows: wherein, represents the optimized vehicle; is the optimized speed of the vehicle at the end of the optimization; is the optimized position of the vehicle at the end of the optimization; auxiliary variable is an integer variable, -1 means the optimized vehicle accelerates after this optimization; 0 means the optimized vehicle keeps the same speed after this optimization; 1 means the optimized vehicle decelerates after this optimization; auxiliary variable is a binary variable, 1 means the optimized vehicle does not leave the target road segment after this optimization; 0 means the optimized vehicle leaves the target road segment. 8.The method of claim 1, wherein, Before using the multi-objective optimization model to calculate the speed change value and lane changing decision of each optimization vehicle, the following steps are included: configuring corresponding following models and lane changing models for all vehicles on the target road section; when using the multi-objective optimization model to calculate the speed change value and lane changing decision of each optimization vehicle, the following steps are included: The artificial driving vehicles on the target road section and the network-connected automatic driving vehicles outside the dynamic emptying area are used as non-optimization vehicles, and the vehicle state information of the non-optimization vehicles at the current time step is input into the corresponding following models and lane changing models respectively to calculate the vehicle state information of the non-optimization vehicles after one optimization step; The vehicle state information of the optimization vehicles at the current time step and the vehicle state information of the non-optimization vehicles after one optimization step are input into the multi-objective optimization model and solved to obtain the speed change value and lane changing decision of each optimization vehicle. 9.The method of claim 1, wherein, When using the speed change value and the optimized lane changing decision to control each optimization vehicle, the following steps are included: If the optimization step of the multi-objective optimization model is greater than the control step, the speed change value is decomposed into each control step to control the speed change of the corresponding optimization vehicle, and the lane changing decision is controlled in the last control step of the corresponding optimization step to control the position change of the corresponding optimization vehicle.

Images