Automatic intersection priority dynamic allocation method based on improved auction algorithm

Abstract

The invention discloses an improved auction algorithm-based automatic intersection priority dynamic allocation method, which comprises the following steps of: assuming that an automatic intersection does not need to be controlled by a signal lamp in an intelligent network connection environment; each lane at the intersection has no fixed lane function attribute, and each lane can allow left turning, straight going and right turning; each vehicle is provided with a wallet, and the vehicle needs to send vehicle basic information and pre-paid fees to the signal control center when the vehicle arrives at the intersection range; establishing a priority distribution model through model hypothesis and model preparation; in a certain time range, priority passing and vehicle payment willingness of emergency vehicles are considered, the passing sequence of the vehicles at the intersection is reasonably distributed through an improved auction algorithm, and the passing priority of the vehicles is determined; according to the invention, the passing order of the vehicles at the intersection can be well arranged, the passing time of the vehicles at the intersection is reduced, the passing right of the emergency vehicles is improved, and the waiting time of the vehicles on the low-flow lane can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of road traffic, in particular to an automatic intersection priority dynamic allocation method based on an improved auction algorithm. Background Art

[0002] In an intelligent connected environment, intersection managers can obtain real-time information about vehicles entering the intersection while simultaneously performing precise control of the vehicles themselves. This allows vehicles to self-organize through intersections, making signal-free intersection control possible. Determining vehicle priority at automated intersections has become a research hotspot in recent years. Current research primarily encompasses rule-based and optimization-based approaches. Dresner proposed a rule based on the order of vehicle arrival (Frist come first serve, FCFS). In the optimization realm, Levin proposed a linear integer programming model based on fixed vehicle speeds and conflict zone occupancy at intersections, solving for vehicle priority at intersections based on this model. Existing methods typically assume equal right of way for vehicles or prioritize emergency vehicles by adding vehicle delay weights to the objective function. These methods lack a flexible response mechanism to dynamic emergency needs.

[0003] In an intelligent connected environment, emergency vehicle access research primarily encompasses route planning, patient information retrieval, accident detection, and driver inattention detection. Research on priority access at intersections focuses on signal preemption and emergency lane reservation. By detecting the arrival of emergency vehicles, intersection signal phases are rationally arranged, and lanes are cleared of vehicles ahead, ensuring smooth and rapid passage for emergency vehicles upon arrival. Currently, research on emergency vehicle priority access at unsignalized intersections is limited, resulting in an inability to address the priority access needs of emergency vehicles, particularly those with multiple emergency vehicles, under unsignalized intersection control conditions.

[0004] Auction theory is a classic economic method widely used in flower sales, art, and antique markets. Carlino first applied auction theory to autonomous intersection vehicle control, allowing users to declare their utility and bid for intersection priority. In auction-based autonomous intersection management (AAIM), drivers have a "wallet agent" that allows drivers to automatically bid on the order of vehicles passing through the intersection based on the value of their time, allowing them to pass through the intersection as early as possible. However, there are three problems with the application of this scheme: first, there is no research on how to handle the passage needs of emergency vehicles and give them sufficiently high priority; second, in this scheme, because the right of way alternates in different directions, it is easy to cause low traffic efficiency; third, entrance lanes with low traffic volume cannot obtain the right of way due to the small number of vehicles, resulting in high traffic delays.

[0005] Based on this, the present invention proposes an automatic intersection priority dynamic allocation algorithm based on an improved auction algorithm, reasonably designs the bidding price and auction process of emergency vehicles, considers the maximum delay of vehicle passage at intersections with low traffic flow, and proposes an automatic driving intersection priority allocation method based on the improved auction algorithm. Summary of the Invention

[0006] To solve the problems existing in the prior art, the purpose of the present invention is to provide a method for dynamic allocation of automatic intersection priority based on an improved auction algorithm. The present invention can better arrange the order of vehicles passing through intersections, improve the right of way for emergency vehicles, and also reduce the waiting time of vehicles in low-traffic lanes.

[0007] To achieve the above-mentioned purpose, the present invention adopts a technical solution: a method for dynamically allocating priorities of automatic intersections based on an improved auction algorithm, comprising the following steps:

[0008] Step 1: Assume that in an intelligent connected environment, automated intersections do not require traffic lights. Each lane at the intersection has independent functional attributes and can allow left turns, straight ahead, and right turns. Each vehicle has a wallet. Upon reaching the intersection, a vehicle must send basic vehicle information and a prepaid fee to bid for intersection priority to the signal control center.

[0009] Step 2: Establish a priority allocation model through model assumptions and model preparation;

[0010] Step 3: Within a certain time frame, considering the priority of emergency vehicles and the willingness to pay of vehicles, the order of vehicles passing at the intersection is reasonably allocated through the improved auction algorithm to determine the vehicle passage priority.

[0011] As a further improvement of the present invention, in step 2, the model assumption specifically includes:

[0012] (1) Communication between vehicles and the intersection control center is smooth and delay-free;

[0013] (2) All vehicles follow the intersection control results and pass through the intersection in the control order;

[0014] (3) The vehicle does not change lanes after entering the intersection control area;

[0015] (4) Assume that drivers are rational in bidding and bid according to their true intentions, and the bidding price does not change with the auction rounds.

[0016] As a further improvement of the present invention, in step 2, the model preparation specifically includes vehicle status, lane-steering conflict function and vehicle bid price.

[0017] As a further improvement of the present invention, the vehicle status is specifically as follows:

[0018] Define the set of vehicles that need to pass through the intersection as V, and the state of vehicle i is s i Including vehicle basic status 1i and bidding status 2i ; Basic vehicle status s 1i Store the lane where the vehicle is located i , vehicle steering κ i , vehicle type τ i and the emergency vehicle mission urgency θ i and task timeliness σ i Basic information, 1i =(l i ,κ i ,τ i ,θ i ,σ i );κ i =0 represents the vehicle going straight, 1 represents turning left, and 2 represents turning right; τ i =0 means the vehicle is a civilian vehicle, 1 means a police car, 2 means an ambulance, and 3 means a fire truck; the urgency of the emergency vehicle mission θ i and task timeliness σ i , when τ i ≠0 is valid, the urgency and timeliness of the task are related to the emergency vehicle's execution of the task; the vehicle bidding status s 2i Stored in the relevant variables of the vehicle bidding process, s 2i =(b it ,r it ,w it ,h it ,k it ), where bit represents the bid price of vehicle i in round t; r it represents the remaining price b of vehicle i in round t it ≤r it ;w it It represents the time that vehicle i has been waiting in the tth round, expressed as the number of rounds the vehicle has been waiting for, that is, δ ik Indicates whether vehicle i passes the kth round, which is a 0-1 variable. If the vehicle fails, then δ ik =1, otherwise, δ k =0;h it k represents the historical credit score of vehicle i in round t. When a vehicle gives way to an emergency vehicle, the credit score increases cumulatively. it It represents the position of vehicle i on the lane at the tth round, that is, the order of the vehicles starting from the stop line, where the order of the first vehicle is 1.

[0019] As a further improvement of the present invention, the lane-steering conflict function is c(l i ,κ j ,l p ,κ q ), if the turn on the lane conflicts with the turn on the lane, the value is 1, otherwise, the value is 0.

[0020] As a further improvement of the present invention, in the vehicle bidding price, the bidding price b of ordinary vehicles it Equal to its bid price, the bid price of the emergency vehicle b it Based on the basic bidding price b0 and the task urgency θ i and task timeliness σ i Jointly decide; normalize the urgency and timeliness of the task, and calculate the bid price of the emergency vehicle by the following formula:

[0021]

[0022] From the formula, we can see that b0<φ it <2b0.

[0023] As a further improvement of the present invention, the step 3 specifically includes the following steps:

[0024] Step 3.1, Initialization: The set of vehicles to be passed V←all vehicles, determine the state function s of vehicle i i , t=0,δ it =0,i∈V;

[0025] Step 3.2: Determine whether there is an emergency vehicle in the vehicle set, τ i ≠0, if yes, go to step 3.3.1; if no, go to step 3.3.2;

[0026] Step 3.3: Determine the winning lane and the winning vehicle i * :

[0027] Step 3.3.1. Calculate the bid value b of the emergency vehicle in round t it , determine the lane where the emergency vehicle with the highest bid value is located as the winning lane The first car in the winning lane is the winning car i * ;Right now

[0028]

[0029] Step 3.3.2: Determine whether there is a vehicle waiting time greater than or equal to the longest waiting time for the first vehicle in the intersection lane. If so, select the vehicle with the highest historical pass score and the timeout as the winner. * , the lane it is in is the winning lane. If the vehicles have the same historical traffic scores, the vehicle is randomly selected; if not, the lane bidding price b is calculated. lt The lane bid price is the sum of the bids of the first k0 vehicles on lane l in the tth round; the lane with the largest bid price is selected as the winning lane The first car in the lane is the winning car i * :

[0030]

[0031] Step 3.4: Determine the set of vehicles P that can pass through this round t :

[0032] According to the winning vehicle i * Turn Construct the current round pass set P t , including the winning vehicle i * , the first set of vehicles P in other lanes that do not conflict with it 1t and the set of vehicles that can pass through the lane P 2t ;P 1t Representation and steering The first vehicle in the other lanes that does not conflict and whose directions in the set do not conflict with each other; P 2t Indicates that the lane and the turn Same passable vehicles, check lanes If the direction of travel of the first l vehicles is the same as that of the first vehicle in the lane, and the direction of travel of any vehicle in front of the vehicle is also The vehicle can obtain the right of way:

[0033] P t ={i *}∪P1t ∪P 2t

[0034]

[0035] Step 3.5: Release vehicles according to the set of passable vehicles, update the vehicle pass variable, update the vehicle waiting time, deduct the vehicle bid fee and update the vehicle residual value. If an emergency vehicle passes, update the credit points of the first vehicle in the other lane. The rule for deducting the vehicle bid fee is the ratio of the bid price of the first l vehicles in the winning lane to the bid price of the lane. Multiply by the second lane bid price b Lt :

[0036] δ it =1←i∈P t

[0037] w it =w it +1,ifδ it =0

[0038]

[0039] Step 3.6: Determine whether all vehicles have passed the intersection. If so, end the process; otherwise, go to step 3.2 at t=t+1.

[0040] The beneficial effects of the present invention are:

[0041] This paper studies automatic intersection priority allocation based on an improved auction algorithm. Case analysis shows that the algorithm proposed in this paper can better arrange the order of vehicles passing through intersections, reducing the time it takes for vehicles to pass through intersections while improving the right of way for emergency vehicles. In addition, the improved auction algorithm can also reduce the waiting time of vehicles in low-traffic lanes and prevent them from waiting too long. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flowchart of the improved auction algorithm in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0044] Example

[0045] A method for dynamic priority allocation at an automatic intersection based on an improved auction algorithm, comprising:

[0046] 1. Problem description:

[0047] In an intelligent connected environment, automatic intersections do not require signal light control. Each lane at the intersection does not have fixed lane function attributes, and each lane can allow left turns, straight driving, and right turns. Each vehicle has a wallet, and when the vehicle arrives at the intersection, it needs to send basic vehicle information and a prepaid fee to bid for the priority of vehicles passing through the intersection to the signal control center. The research problem of this embodiment can be described as: within a certain time range, considering the priority passage of emergency vehicles and the willingness to pay of vehicles, by improving the auction algorithm, the order of vehicles passing at the intersection is reasonably allocated, the priority of vehicles passing is determined, and emergency vehicles pass through the intersection as quickly as possible, while also improving the release rate of the intersection.

[0048] 2. Priority allocation model:

[0049] 2.1 Model Assumptions:

[0050] (1) Communication between vehicles and the intersection control center is smooth and delay-free;

[0051] (2) All vehicles follow the intersection control results and pass through the intersection in the control order;

[0052] (3) The vehicle does not change lanes after entering the intersection control area;

[0053] (4) Assume that drivers are rational in bidding and bid according to their true intentions, and the bidding price does not change with the auction rounds.

[0054] 2.2 Model preparation:

[0055] (1) Vehicle status:

[0056] Define the set of vehicles that need to pass through the intersection as V, and the state of vehicle i is s i Including vehicle basic status 1i and bidding status 2i . Basic vehicle status 1i Lane where the main storage vehicle is located l i , vehicle steering κ i , vehicle type τ i and the emergency vehicle mission urgency θ i and task timeliness σ i Basic information such as 1i =(l i ,κ i ,τ i ,θ i ,σ i ). κ i =0 represents the vehicle going straight, 1 represents turning left, and 2 represents turning right; τ i =0 means the vehicle is a civilian vehicle, 1 means a police car, 2 means an ambulance, and 3 means a fire truck. Emergency vehicle mission urgency θi and task timeliness σ i , when τ i Valid when ≠0, the urgency and timeliness of the task are related to the emergency vehicle's execution of the task, which will be detailed in the vehicle bidding price. Vehicle bidding status 2i Mainly stores the relevant variables in the vehicle bidding process, s 2i =(b it ,r it ,w it ,h it ,k it ), where b it represents the bid price of vehicle i in round t; r it represents the remaining price b of vehicle i in round t it ≤r it ;w it It represents the time that vehicle i has been waiting in the tth round, expressed as the number of rounds the vehicle has been waiting for, that is, δ ik Indicates whether vehicle i passes the kth round, which is a 0-1 variable. If the vehicle fails, then δ ik =1, otherwise, δ k =0.h it k represents the historical credit score of vehicle i in round t. When a vehicle gives way to an emergency vehicle, the credit score increases cumulatively. it It represents the position of vehicle i on the lane at the tth round, that is, the order of the vehicles starting from the stop line, where the order of the first vehicle is 1.

[0057] (2) Lane-turn conflict function:

[0058] Define lane-steering conflict function c(l i ,κ j ,l p ,κ q ), if the turn on the lane conflicts with the turn on the lane, the value is 1, otherwise, the value is 0. Taking one lane per entrance as an example, the lane-turn conflict function of the intersection is shown in Table 1.

[0059] Table 1 Lane-turn conflicts at intersections

[0060]

[0061]

[0062] (3) Vehicle bid price:

[0063] Bid price of ordinary vehicle b it Equal to its bid price, the bid price of the emergency vehicle b itBased on the basic bidding price b0 and the task urgency θ i and task timeliness σ i Joint decision. Urgency θ i It measures the priority of emergency tasks and reflects the severity of the consequences if no immediate response is given. i The value range is [0,10], refer to the table below. Task timeliness σ i The remaining time to the task deadline or optimal processing time reflects the urgency of the task. To prevent a vehicle with extremely high urgency or extremely high task timeliness from completely dominating the bidding results, urgency is calculated using a logarithmic function and task timeliness is calculated using an exponential function. By normalizing urgency and task timeliness, the bid price for an emergency vehicle is calculated using the following formula. As can be seen from the formula, b0 < φ it <2b0.

[0064]

[0065] Table 2 Urgency θ i Classification table

[0066]

[0067] 2.3. Improved auction algorithm:

[0068] The steps of improving the auction algorithm are as follows. The detailed process is as follows Figure 1 As shown:

[0069] Step 1: Initialization. The set of vehicles to be passed V←all vehicles, determine the state function s of vehicle i i , t=0,δ it =0,i∈V.

[0070] Step 2: Determine whether there is an emergency vehicle in the vehicle set, τ i ≠0, if yes, go to step 3.1; if not, go to step 3.2.

[0071] Step 3: Determine the winning lane and the winning vehicle i * :

[0072] Step 3.1: Calculate the bid value b of the emergency vehicle in round t it , determine the lane where the emergency vehicle with the highest bid value is located as the winning lane The first car in the winning lane is the winning car i * .Right now

[0073]

[0074] Step 3.2: Determine whether the first vehicle in the intersection lane has a waiting time greater than or equal to the maximum waiting time. If so, select the vehicle with the highest historical pass score and the timeout as the winning vehicle i * , the lane it is in is the winning lane. If the vehicles have the same historical traffic scores, the vehicle is randomly selected; if not, the lane bidding price b is calculated. lt The lane bid price is the sum of the bids of the first k0 vehicles on lane l in the tth round; the lane with the largest bid price is selected as the winning lane The first car in the lane is the winning car i * :

[0075]

[0076] Step 4: Determine the set of vehicles P that can pass through this round t .

[0077] According to the winning vehicle i * Turn Construct the current round pass set P t , including the winning vehicle i * , the first set of vehicles P in other lanes that do not conflict with it 1t and the set of vehicles that can pass through the lane P 2t .P 1t Representation and steering The first vehicle in the set that does not conflict with the other lanes and whose directions do not conflict with each other. 2t Indicates that the lane and the turn Same passable vehicles, check lanes The direction of travel of the first l vehicles, if it is in the same lane as lane l 0t The direction of travel of the first car is the same as that of the first car, and the direction of travel of any car in front of the car is also The vehicle can obtain the right of way.

[0078] P t ={i *}∪P 1t ∪P 2t

[0079]

[0080] Step 5: Release vehicles according to the set of passable vehicles, update the vehicle pass variable, update the vehicle waiting time, deduct the vehicle bid fee and update the vehicle residual value. If an emergency vehicle passes, update the credit points of the first vehicle in the other lane. The rule for deducting the vehicle bid fee is the ratio of the bid price of the first l vehicles in the winning lane to the bid price of the lane. Multiply by the second lane bid price b Lt .

[0081] δ it =1←i∈P t

[0082] w it =w it +1,ifδ it =0

[0083] r it =r it -b it ,ifδ it =1

[0084]

[0085] Step 6: Determine whether all vehicles have passed the intersection. If so, end the process; otherwise, go to step 2 at t=t+1.

[0086] 3. Numerical simulation verification:

[0087] To verify the effectiveness of the auction algorithm proposed in this example, the auction algorithm and the original auction algorithm were compared in terms of vehicle release speed and vehicle release speed on secondary roads. The vehicle release speed was expressed as the number of rounds entering the intersection. The vehicle release speed performance was compared under a uniform scenario (consistent vehicle numbers in all lanes), a primary-secondary scenario (low vehicle count in one entry lane), and scenarios with a single emergency vehicle and multiple emergency vehicles exiting.

[0088] 3.1. Uniform scene:

[0089] Each lane at the intersection was set at 8 vehicles, for a total of 32 vehicles. The ratio of straight-through to left-turn traffic was 1:1. Vehicle turns were randomly generated, with a uniform initial price of 20. Vehicle bids were integers between [2 and 5], randomly generated using the random module. The number of rounds for releasing vehicles was compared between the original auction method and the current auction method. Three simulations were conducted to simulate the arrival of randomly generated vehicles and their bid prices. The information for the first randomly generated vehicle is shown in Table 3, and the number of rounds the vehicles passed through the intersection is shown in Table 4. During these three vehicle releases, the average intersection vehicle passing speed decreased from 20.67 rounds to 14.67 rounds, a 29.0% improvement.

[0090] Table 3 The first random generation of vehicle information

[0091]

[0092] Table 4 Comparison of the number of passes of the first randomly generated vehicle at the intersection

[0093]

[0094] 3.2. Primary and secondary scenes:

[0095] Under the primary-secondary state, the north, east, and south entrance lanes are set at 8 vehicles per lane, and the west entrance lane is set at 4 vehicles per lane, for a total of 28 vehicles at the intersection. The ratio of straights to left turns is 1:1. Vehicle turns are randomly generated, the initial price is uniformly set at 20, and vehicle bids are integers between [2 and 5], randomly generated using the random module. The number of rounds for releasing vehicles is compared between the original auction method and the current auction method. Three simulations of randomly generated vehicle arrivals and vehicle bids are conducted. During these three vehicle releases, the average vehicle speed at the intersection decreased from 18.3 rounds to 11.7 rounds, a 36.3% improvement. Taking the first random generation as an example, the number of rounds for vehicles 24, 25, 26, and 27 on the low-traffic entrance lanes increased from 6, 11, 16, and 18 to 6, 8, 10, and 10, respectively, an average increase of 33.3%. This demonstrates that the improved auction algorithm can accelerate the release of vehicles on secondary arterial roads.

[0096] 3.3. Single emergency vehicle passage:

[0097] Each lane at the intersection is set to 8 vehicles, the total number of vehicles at the intersection is 32, the straight-through left-turn ratio is 1:1, the vehicle direction, vehicle bidding price, and initial price are the same as the first randomly generated vehicle direction in the uniform scenario. The 14th vehicle is adjusted to an emergency vehicle (fire truck, lane 1, vehicle direction is 1, the fire truck mission is a large fire rescue with no trapped personnel, and the urgency level θ i The value is 8. Task timeliness σ i The rescue mission limit is 8 minutes. max The results show that the automatic intersection first releases vehicles in the emergency vehicle lane (lane 1) until emergency vehicle 14 passes through the intersection. The emergency vehicle has the maximum right of way, meeting the emergency vehicle priority requirement. The emergency vehicle pass round number increases from the 8th to the 5th, an increase of 37.5%. In the improved algorithm, the intersection releases vehicles in the emergency vehicle lane until the emergency vehicle passes.

[0098] Table 5 The first randomly generated number of vehicle passes through the intersection

[0099]

[0100]

[0101] 3.4、Multiple emergency vehicles pass:

[0102] Multiple emergency vehicle input Based on the single emergency vehicle input, the fourth vehicle in lane 3 is modified into an emergency vehicle (vehicle number 27, ambulance, whose mission is to transport patients with sudden cardiac arrest, with an emergency level of θ i The value is 9. Task timeliness σ i 3 minutes, the upper limit of medical tasks is σ max The time taken is 15 minutes, the urgency coefficient is 0.960, the time sensitivity coefficient is 0.234, and the comprehensive bid price is 1.224b0 (assuming b0 is 20 and the comprehensive bid price is 25). Compared to fire trucks, ambulances have higher bid prices. Based on this input, the traffic flow at the intersection under the improved auction algorithm was determined. The vehicle release is shown in Table 6. The results show that the automated intersection first releases vehicles in the ambulance's lane (lane 3) until the ambulance passes through the intersection, and then releases vehicles in the fire truck's lane (lane 1) until the fire truck passes through the intersection. This release process meets the traffic flow requirements of multiple emergency vehicles. The vehicle in front of the emergency vehicle in lane 3 is indicated by an "*" in the upper right corner, and the vehicle in front of the emergency vehicle in lane 1 is indicated by a "#" in the upper right corner.

[0103] Table 6 The first randomly generated number of vehicle passes through the intersection

[0104]

[0105]

[0106] The above-described embodiments merely represent specific implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, and all such variations and improvements fall within the scope of protection of the present invention.

Claims

1. A method for dynamic allocation of automatic intersection priority based on an improved auction algorithm, characterized in that: The following steps are involved: Step 1: Assume that in an intelligent connected environment, automated intersections do not require traffic lights. Each lane at the intersection has independent functional attributes and can allow left turns, straight ahead, and right turns. Each vehicle has a wallet. Upon reaching the intersection, a vehicle must send basic vehicle information and a prepaid fee to bid for intersection priority to the signal control center. Step 2: Establish a priority allocation model through model assumptions and model preparation; Step 3: Within a certain time frame, considering the priority of emergency vehicles and the willingness to pay of vehicles, the order of vehicles passing at the intersection is reasonably allocated through the improved auction algorithm to determine the vehicle passage priority.

2. The method for dynamic allocation of automatic intersection priority based on improved auction algorithm according to claim 1, characterized in that: In step 2, the model assumptions specifically include: (1) Communication between vehicles and the intersection control center is smooth and delay-free; (2) All vehicles follow the intersection control results and pass through the intersection in the control order; (3) The vehicle does not change lanes after entering the intersection control area; (4) Assume that drivers are rational in bidding and bid according to their true intentions, and the bidding price does not change with the auction rounds.

3. The method for dynamic allocation of automatic intersection priority based on improved auction algorithm according to claim 1, characterized in that: In step 2, the model preparation specifically includes vehicle status, lane-steering conflict function and vehicle bid price.

4. The method for dynamic allocation of automatic intersection priority based on improved auction algorithm according to claim 3, characterized in that: The vehicle status is as follows: Define the set of vehicles that need to pass through the intersection as V, and the state of vehicle i is s i Including vehicle basic status 1i and bidding status 2i ; Basic vehicle status s 1i Store the lane where the vehicle is located i , vehicle steering κ i , vehicle type τ i and the emergency vehicle mission urgency θ i and task timeliness σ i Basic information, 1i =(l i ,κ i ,τ i ,θ i ,σ i );κ i =0 represents the vehicle going straight, 1 represents turning left, and 2 represents turning right; τ i =0 means the vehicle is a civilian vehicle, 1 means a police car, 2 means an ambulance, and 3 means a fire truck; the urgency of the emergency vehicle mission θ i and task timeliness σ i , when τ i ≠0 is valid, the urgency and timeliness of the task are related to the emergency vehicle's execution of the task; the vehicle bidding status s 2i Stored in the relevant variables of the vehicle bidding process, s 2i =(b it ,r it ,w it ,h it ,k it ), where b it represents the bid price of vehicle i in round t; r it represents the remaining price b of vehicle i in round t it ≤r it ;w it It represents the time that vehicle i has been waiting in the tth round, expressed as the number of rounds the vehicle has been waiting for, that is, δ ik Indicates whether vehicle i passes the kth round, which is a 0-1 variable. If the vehicle fails, then δ ik =1, otherwise, δ k =0;h it k represents the historical credit score of vehicle i in round t. When a vehicle gives way to an emergency vehicle, the credit score increases cumulatively. it It represents the position of vehicle i on the lane at the tth round, that is, the order of the vehicles starting from the stop line, where the order of the first vehicle is 1.

5. The method for dynamic allocation of automatic intersection priority based on improved auction algorithm according to claim 4, characterized in that: The lane-steering conflict function is c(l i ,κ j ,l p ,κ q ), if the turn on the lane conflicts with the turn on the lane, the value is 1, otherwise, the value is 0.

6. The method for dynamic allocation of automatic intersection priority based on improved auction algorithm according to claim 5, characterized in that: Among the vehicle bidding prices, the bidding price of ordinary vehicles is b it Equal to its bid price, the bid price of the emergency vehicle b it Based on the basic bidding price b0 and the task urgency θ i and task timeliness σ i Jointly decide; normalize the urgency and timeliness of the task, and calculate the bid price of the emergency vehicle by the following formula: From the formula, we can see that b0<φ it <2b0.

7. The method for dynamic allocation of automatic intersection priority based on improved auction algorithm according to claim 6, characterized in that: The step 3 specifically includes the following steps: Step 3.1, Initialization: The set of vehicles to be passed V←all vehicles, determine the state function s of vehicle i i , t=0,δ it =0,i∈V; Step 3.2: Determine whether there is an emergency vehicle in the vehicle set, τ i ≠0, if yes, go to step 3.3.1; if no, go to step 3.3.2; Step 3.3: Determine the winning lane and the winning vehicle i * : Step 3.3.

1. Calculate the bid value b of the emergency vehicle in round t it , determine the lane where the emergency vehicle with the highest bid value is located as the winning lane The first car in the winning lane is the winning car i * ;Right now Step 3.3.2: Determine whether there is a vehicle waiting time greater than or equal to the longest waiting time for the first vehicle in the intersection lane. If so, select the vehicle with the highest historical pass score and the timeout as the winner. * , the lane it is in is the winning lane. If the vehicles have the same historical traffic scores, the vehicle is randomly selected; if not, the lane bidding price b is calculated. lt The lane bid price is the sum of the bids of the first k0 vehicles on lane l in the tth round; the lane with the largest bid price is selected as the winning lane The first car in the lane is the winning car i * : Step 3.4: Determine the set of vehicles P that can pass through this round t : According to the winning vehicle i * Turn Construct the current round pass set P t , including the winning vehicle i * , the first set of vehicles P in other lanes that do not conflict with it 1t and the set of vehicles that can pass through the lane P 2t ;P 1t Representation and steering The first vehicle in the other lanes that does not conflict and whose directions in the set do not conflict with each other; P 2t Indicates that the lane and the turn Same passable vehicles, check lane l t * If the direction of travel of the first l vehicles is the same as that of the first vehicle in the lane, and the direction of travel of any vehicle in front of the vehicle is also The vehicle can obtain the right of way: P t ={i * }∪P 1t ∪P 2t Step 3.5: Release vehicles according to the set of passable vehicles, update the vehicle pass variable, update the vehicle waiting time, deduct the vehicle bid fee and update the vehicle residual value. If an emergency vehicle passes, update the credit points of the first vehicle in the other lane. The rule for deducting the vehicle bid fee is the ratio of the bid price of the first l vehicles in the winning lane to the bid price of the lane. Multiply by the second lane bid price b Lt : d it =1←i∈P t In it =in it +1,ifδ it =0 Step 3.6: Determine whether all vehicles have passed the intersection. If so, end the process; otherwise, go to step 3.2 at t=t+1.

Images