A signal intersection lane allocation method and device considering random traffic flow

Abstract

The present invention discloses a signal intersection lane allocation method and device considering random traffic flow, wherein the method comprises: obtaining random traffic flow; performing random programming processing on the random traffic flow, constructing a model with minimum traffic flow delay as the objective function, and obtaining a two-stage random programming model; performing decoupling processing on the two-stage random programming model according to the service level reliability, and obtaining a first-stage decoupling model and a second-stage decoupling model; performing optimization iterative calculation processing on the first-stage decoupling model and the second-stage decoupling model according to the gradient descent algorithm, and obtaining a service level reliability index combination; performing solution processing on the first-stage decoupling model according to the service level reliability index combination, and obtaining a signal intersection lane allocation result. The embodiment of the present invention can improve the robustness of lane allocation, reduce the computational complexity and computational cost, and can be widely used in the field of road traffic management technology.

Description

Technical Field

[0001] The present invention relates to the technical field of road traffic management, and in particular to a signal intersection lane allocation method and device considering random vehicle flow. Background Art

[0002] With the rapid development of social economy, people's travel needs are increasing, and the mismatch between road supply and travel demand is becoming more and more serious. As the center of traffic flow gathering and evacuation, intersections often become bottleneck sections in the traffic network, which has a key impact on delays. The existing intersection lane allocation method assumes that the arrival of traffic at the intersection is a fixed value and allocates lanes according to signal control. However, due to the volatility and randomness of traffic demand, the number of vehicles arriving in each cycle is usually not the same, and the related methods have problems of poor interpretability and high computational cost. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a method and device for allocating lanes at a signalized intersection taking into account random traffic flow, so as to improve the robustness of lane allocation and reduce the computational cost.

[0004] In one aspect, the present invention provides a method for allocating lanes at a signalized intersection taking into account random traffic flow, comprising:

[0005] Get random traffic flow;

[0006] Performing random programming on the random traffic flow, building a model with minimum traffic delay as the objective function, and obtaining a two-stage random programming model;

[0007] Decoupling the two-stage stochastic programming model according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model;

[0008] The first-stage decoupling model and the second-stage decoupling model are optimized and iterated according to a gradient descent algorithm to obtain a service level reliability indicator combination;

[0009] The first-stage decoupling model is solved according to the service level reliability index combination to obtain a signalized intersection lane allocation result.

[0010] Optionally, the random traffic flow is subjected to random programming processing, and a model is constructed with minimum traffic delay as the objective function to obtain a two-stage random programming model, including:

[0011] Performing random scene division processing on the random traffic flow to obtain a random scene set;

[0012] Calculating the minimum delay of traffic flow according to the random scenario set to obtain the objective function;

[0013] The model is constructed according to the objective function to obtain a two-stage stochastic programming model.

[0014] Optionally, the calculating the minimum delay of traffic flow according to the random scene set to obtain the objective function includes:

[0015] The objective function includes a first-stage objective function and a second-stage objective function;

[0016] Calculating the expected minimum delay of traffic flow for all scenarios in the random scenario set to obtain the objective function of the first stage;

[0017] The first-stage objective function is solved to obtain a lane allocation scheme, and traffic delay is calculated for any one of the random scene sets according to the lane allocation scheme to obtain the second-stage objective function.

[0018] Optionally, performing traffic delay calculation for any one of the random scene sets according to the lane allocation scheme to obtain the second-stage objective function includes:

[0019] Performing traffic cycle division processing on any one scene in the random scene set according to the lane allocation scheme to obtain a traffic cycle set;

[0020] Performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay;

[0021] The second-stage objective function is obtained according to the scenario lane delay and the scenario traffic flow.

[0022] Optionally, performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay includes:

[0023] Obtaining the cycle length, traffic arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles of each cycle in the traffic cycle set;

[0024] The cycle length, vehicle arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles are calculated using a lane delay calculation formula to obtain the scenario lane delay.

[0025] Optionally, the two-stage stochastic programming model is decoupled according to the service level reliability to obtain a first-stage decoupled model and a second-stage decoupled model, including:

[0026] Initialize service level reliability indicator;

[0027] The reserve capacity of the first stage of the two-stage stochastic programming model is calculated by using the service level reliability index to obtain a first-stage decoupling model;

[0028] The traffic delay calculation process of the second stage of the two-stage stochastic programming model is performed according to the first-stage decoupling model to obtain a second-stage decoupling model.

[0029] Optionally, the optimizing iterative calculation processing of the first-stage decoupling model and the second-stage decoupling model according to the gradient descent algorithm to obtain the service level reliability indicator combination includes:

[0030] Initialize the service level reliability set;

[0031] Inputting the service level reliability set into the first stage decoupling model to perform lane allocation calculation processing to obtain allocated lanes;

[0032] performing expected total delay calculation processing on the second-stage decoupling model according to the assigned lanes to obtain an expected total delay difference;

[0033] The service level reliability set is updated according to the total delay expectation difference, and the process returns to the step of inputting the service level reliability set into the first-stage decoupling model for lane allocation calculation processing to obtain the lane allocation step, until the total delay expectation difference is less than a preset threshold, and the service level reliability set is determined as the service level reliability indicator combination.

[0034] On the other hand, an embodiment of the present invention further provides a signalized intersection lane allocation device considering random vehicle flow, comprising:

[0035] The first module is used to obtain random traffic flow;

[0036] The second module is used to perform random programming on the random traffic flow, build a model with the minimum delay of the traffic flow as the objective function, and obtain a two-stage random programming model;

[0037] The third module is used to decouple the two-stage stochastic programming model according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model;

[0038] The fourth module is used to optimize and iterate the first-stage decoupling model and the second-stage decoupling model according to the gradient descent algorithm to obtain a service level reliability indicator combination;

[0039] The fifth module is used to solve the first-stage decoupling model according to the service level reliability indicator combination to obtain the signalized intersection lane allocation result.

[0040] On the other hand, an embodiment of the present invention further discloses an electronic device, including a processor and a memory;

[0041] The memory is used to store programs;

[0042] The processor executes the program to implement the method described above.

[0043] On the other hand, an embodiment of the present invention further discloses a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the method described above.

[0044] On the other hand, an embodiment of the present invention further discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above method.

[0045] Compared with the prior art, the present invention adopts the above technical scheme and has the following technical effects: the embodiment of the present invention performs random programming on the random traffic flow, constructs a model with the minimum delay of the traffic flow as the objective function, and obtains a two-stage random programming model; it can be applied to the scenario of random traffic flow arrival, and constructs a model with minimizing delay as the optimization goal, which effectively reduces the expected delay of traffic flow at the intersection and improves the robustness of the lane allocation scheme; in addition, the embodiment of the present invention also introduces the concept of service level reliability to decouple the model, and transforms the problem into solving the service level reliability combination that minimizes the expected total delay, thereby reducing the computational complexity and improving the solution efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 is a flow chart of a signalized intersection lane allocation method considering random traffic flow provided by an embodiment of the present application;

[0048] Figure 2 is a lane delay calculation illustration diagram provided in an embodiment of the present application;

[0049] Figure 3 It is a coupling relationship illustration diagram of a two-stage stochastic programming model provided in an embodiment of the present application;

[0050] Figure 4 It is a flow chart of solving a decoupled model provided in an embodiment of the present application;

[0051] Figure 5 This is a schematic diagram of a random vehicle flow entrance and exit provided in an embodiment of the present application;

[0052] Figure 6 It is a schematic diagram of lane allocation adjustment changes provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] In related technologies, intersection optimization is mainly based on signal control optimization, that is, the travel time and travel order of conflicting traffic flows are controlled by optimizing signal timing. As an important means of allocating spatial resources at intersections, lane allocation at intersections can be coordinated with signal control to further improve the operating efficiency of intersections, so it has become one of the keys to studying intersection optimization. Most methods for coordinated optimization of lane function allocation and signal control assume that the arrival of traffic at intersections is a fixed value. In fact, traffic demand is volatile and random, and the number of vehicles arriving in each cycle is usually different. In signal timing optimization, in order to cope with this volatility and randomness, intersections can often be optimized by adjusting signal timing parameters in real time. When co-optimizing lane allocation and signal timing, considering the potential safety hazards and additional delays caused by frequent real-time adjustments to lane functions, it is often expected to find a fixed lane allocation scheme over a long period of time, and under this scheme, adjust the signal timing in real time to obtain a control scheme that minimizes delays. At present, there is little research on the real-time optimization coordination of lane function allocation and traffic signal timing, especially research with minimum delay as the objective function. Existing methods also have problems such as poor interpretability and high computational cost.

[0055] In view of this, refer to Figure 1 , an embodiment of the present invention provides a signalized intersection lane allocation method considering random vehicle flow, comprising:

[0056] S101, obtaining random traffic flow;

[0057] S102, performing random programming processing on the random traffic flow, constructing a model with minimum traffic delay as the objective function, and obtaining a two-stage random programming model;

[0058] S103, decoupling the two-stage stochastic programming model according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model;

[0059] S104, performing iterative optimization calculation processing on the first-stage decoupling model and the second-stage decoupling model according to a gradient descent algorithm to obtain a service level reliability indicator combination;

[0060] S105. Solve the first-stage decoupling model according to the service level reliability index combination to obtain a signalized intersection lane allocation result.

[0061] In an embodiment of the present invention, it is applied to intersection lane allocation in a random traffic flow scenario. The embodiment of the present invention first obtains random traffic flow, performs random planning on the random traffic flow, and constructs a two-stage random programming model with minimum delay as the objective function. The two-stage random programming model includes a first-stage model and a second-stage model. According to the definition of the problem, the two stages of the two-stage random programming problem need to meet the following requirements: In the first stage, according to the first-stage model, a lane allocation scheme that minimizes the expected total delay in all possible scenarios is obtained, and the scheme remains unchanged during the time period covered by all scenarios; in the second stage, according to the second-stage model, a signal timing scheme that minimizes the total delay in a specific scenario under a fixed lane allocation scheme is obtained, thereby adapting to the randomness of traffic arrival.

[0062] Then, taking service level reliability as the decoupling indicator, the highly coupled two-stage stochastic programming model is decoupled, and the decoupled first and second stage models are constructed. The optimal reliability indicator combination is solved by the gradient descent algorithm, and the lane allocation scheme corresponding to this set of reliability indicator combination is obtained, that is, the lane allocation result of the signalized intersection.

[0063] It should be noted that in the process of constructing the two-stage stochastic programming model, the constraints of the two-stage stochastic programming model need to be set:

[0064] (1) Minimum lane right-of-way constraint: This means that each lane must have at least one direction of right-of-way (or turning function), which can be expressed as

[0065]

[0066] Among them, δ i，j，k Indicates whether the driving direction j of the k-th lane of entrance i is permitted. When its value is 1, it is permitted, and when its value is 0, it is not permitted.

[0067] (2) Adjacent lane function constraints: This means ensuring that the steering functions of adjacent lanes do not conflict with each other, which can be expressed as:

[0068] 1-δ t，j，k+1 ≥δ i，m，k ;

[0069] Among them, i=0, 1,...,i max ; j = 0, 1, ... j max -1; m = j, j + 1, ... j max and k = 0, 1, ... k max .

[0070] (3) Constraint on the number of inlet and outlet channels: that is, to ensure that the number of outlet channels is not less than the number of inlet channels, which can be expressed as

[0071]

[0072] Among them, A T(i，j) Represents the lane number of the exit lane with sequence number T(i, j), where T(i, j) refers to the exit lane sequence number to which the traffic starting from entrance i and traveling in direction j is heading.

[0073] (4) Cycle duration constraint: This means ensuring that the cycle duration is within the specified value range, which can be expressed as

[0074]

[0075] Among them, ξ is the inverse of the signal cycle length; c min and c max They are the minimum and maximum values ​​of the cycle duration, respectively, and are usually determined based on the actual scenario.

[0076] (5) Green light start time and duration constraints: that is, to ensure that the green light start time and duration are within the specified value range (determined according to the actual scenario). The green light start time constraint can be expressed as

[0077]

[0078] The green light duration constraint can be expressed as

[0079]

[0080] Among them, θ i represents the green light start time of import i, φ i represents the green light duration of import i, g min and g in They represent the minimum value of the phase green light time and the phase green light interval time respectively.

[0081] (6) Shared lane signal control constraint: This means that when a lane allows two or more traffic flows to turn, that is, the lane is a shared lane, then they must be controlled by the same signal phase, which can be expressed as

[0082] M(1-δ i，j，k )≥Θ i，k -θ i ≥-M(1-δ i，j，k );

[0083] M(1-δ i，j，k )≥Φ i，k -φ i ≥-M(1-δ i，j，k );

[0084] Among them, Θ i，k represents the green light start time of the kth lane of the import i, Φ i，k represents the green light duration of the k-th lane at the import i, and M is a sufficiently large positive number.

[0085] (7) Flow conservation constraint: This means ensuring that the allocated flow can meet traffic demand, which can be expressed as

[0086]

[0087] Among them, Q i，j represents the traffic demand for import i in direction j, which is the input quantity; q i，j，k It indicates the traffic volume with direction j assigned to the k-th lane of entrance i.

[0088] (8) Prohibited traffic constraint: This means that when a lane does not have the right of way in a certain direction, traffic in that direction cannot be allocated to that lane. It can be expressed as

[0089] M·δ i，j，k ≥q i，j，k ;

[0090] (9) Lane equal flow ratio constraint: This means that when adjacent lanes have the same turning function, their flow ratios are equal, which can be expressed as

[0091]

[0092] Among them, r i，j，k represents the turning radius of the traffic flow in the k-th lane of the entrance i with the driving direction j, It represents the straight saturation flow rate of the k-th lane of the entrance i. The specific value is usually determined according to the road requirements in the actual scenario.

[0093] As a further preferred implementation, the random traffic flow is subjected to random programming, and a model is constructed with minimum traffic delay as the objective function to obtain a two-stage random programming model, including:

[0094] Performing random scene division processing on the random traffic flow to obtain a random scene set;

[0095] Calculating the minimum delay of traffic flow according to the random scenario set to obtain the objective function;

[0096] The model is constructed according to the objective function to obtain a two-stage stochastic programming model.

[0097] In the embodiment of the present invention, the random traffic flow is divided into random scenarios to obtain a random scenario set. When the random traffic demand meets a specific continuous probability distribution, the expectation of the total delay in all scenarios can be expressed in integral form, so the two-stage random programming model It can be expressed as:

[0098]

[0099] Where Dr is the delay obtained in the second stage, and P is the probability of this situation occurring. The objective function D in the first stage represents the objective function Dr in the second stage. k The expectation of the total delay in each scenario.

[0100] However, considering that this form of solution is difficult to obtain, in this type of problem, scenario simulation is generally used to estimate the expectation. Assume that the number of scenarios to be generated is l. The scenario set can be represented as S = {1, 2, ..., l}. The probability of each scenario appearing is equal. Further expressed as:

[0101]

[0102] Where D represents the objective function of the first stage, represents the minimum expected total delay of traffic flow when only lane allocation is adjusted, E represents expectation, Dr represents the second stage of the random programming model, G represents the signal control parameter of random traffic flow, Δ represents the lane allocation scheme, Q represents the traffic demand of random traffic flow, κ represents the κth scenario in the random scenario set, κ max represents the total number of random scene sets, P κ represents the probability of occurrence of the κth scene, Q κ represents the traffic demand of the κth scenario, Dr k represents the second stage objective function, G κ represents the signal control parameters of the κth scene.

[0103] As a further preferred implementation, the minimum traffic delay calculation is performed according to the random scene set to obtain the objective function, including:

[0104] The objective function includes a first-stage objective function and a second-stage objective function;

[0105] Calculating the expected minimum delay of traffic flow for all scenarios in the random scenario set to obtain the objective function of the first stage;

[0106] The first-stage objective function is solved to obtain a lane allocation scheme, and traffic delay is calculated for any one of the random scene sets according to the lane allocation scheme to obtain the second-stage objective function.

[0107] In an embodiment of the present invention, in a scenario where random traffic arrives, a two-stage random programming model with minimum delay as the objective function is constructed. In the first stage, the two-stage random programming model solves a fixed lane allocation scheme that minimizes delays in all scenarios, so the first stage objective function is the expected minimum delay of traffic calculated for all scenarios in the random scenario set; in the second stage, the model solves a signal timing scheme that minimizes delays in each specific scenario under the fixed lane allocation scheme, and further finds the minimum value of the expected total delay of all scenarios under the lane allocation scheme, so the second stage objective function is the traffic delay calculated for any one scenario in the random scenario set according to the fixed lane allocation scheme solved in the first stage.

[0108] As a further preferred implementation, the traffic delay calculation is performed on any one of the random scene sets according to the lane allocation scheme to obtain the second stage objective function, including:

[0109] Performing traffic cycle division processing on any one scene in the random scene set according to the lane allocation scheme to obtain a traffic cycle set;

[0110] Performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay;

[0111] The second-stage objective function is obtained according to the scenario lane delay and the scenario traffic flow.

[0112] In the embodiment of the present invention, the second stage objective function Dr κ It can be expressed as:

[0113]

[0114] In the formula, i represents the number of the entrance lane of the intersection, j represents the driving direction of the random traffic flow, k represents the number of the lane, n represents the traffic cycle, q represents the traffic volume, and d represents the lane delay. Indicates the minimum delay in a specific scenario that can be obtained by adjusting signal timing alone.

[0115] As a further preferred implementation, the lane delay calculation process is performed on the traffic cycle set to obtain the scene lane delay, including:

[0116] Obtaining the cycle length, traffic arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles of each cycle in the traffic cycle set;

[0117] The cycle length, vehicle arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles are calculated using a lane delay calculation formula to obtain the scenario lane delay.

[0118] In the embodiment of the present invention, the lane delay calculation formula is as follows:

[0119]

[0120] The constraint condition in formula (1) is

[0121] The constraint condition in formula (2) is

[0122] Among them, C represents the cycle length, q represents the traffic arrival rate, Φ represents the green light duration of the lane, and s represents the saturation flow rate, such as represents the saturated flow rate on the kth lane of the ith entrance in the nth cycle of the κth scenario.

[0123] And H represents the number of remaining queued vehicles, expressed as:

[0124]

[0125]

[0126] In the formula, It represents the number of remaining queued vehicles on the kth lane of the i-th entrance in the n-th cycle of the κ-th scenario.

[0127] It should be noted that the lane delay and the number of remaining queued vehicles are derived as follows: Figure 2 As shown in the figure, the delay values ​​of undersaturation and oversaturation can be calculated by calculating the area of ​​the shaded part in the figure respectively. The number of remaining vehicles is obtained by calculating the relationship between the upper and lower parts of the shaded part.

[0128] As a further preferred implementation, the two-stage stochastic programming model is decoupled according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model, including:

[0129] Initialize service level reliability indicator;

[0130] The reserve capacity of the first stage of the two-stage stochastic programming model is calculated by using the service level reliability index to obtain a first-stage decoupling model;

[0131] The traffic delay calculation process of the second stage of the two-stage stochastic programming model is performed according to the first-stage decoupling model to obtain a second-stage decoupling model.

[0132] In the embodiment of the present invention, Figure 3 As mentioned above, in the two-stage stochastic programming model, solving the second-stage problem requires obtaining in advance the lane function allocation plan that minimizes the total delay in the first stage. The objective function of the second stage, as a component of the objective function of the first stage, affects the result of the lane function allocation plan. These two problems are coupled, and solving the delay is a nonlinear programming problem, which will significantly increase the difficulty of solving the problem.

[0133] In order to decouple the above two-stage stochastic programming model, the concept of service level reliability is introduced. Service level reliability refers to the probability that a certain direction of the entrance road of the intersection can meet a specific service level. The numerical value is equivalent to the proportion of all possible scenarios where traffic demand can be met at a specific service level. Among them, the division of service level levels in a certain direction of the entrance road is consistent with the division of service levels of national highway sections.

[0134] Among them, the service level reliability index ρ of the import channel i direction j i，j It can be expressed as:

[0135]

[0136] Among them, it means P r The probability that the condition in brackets is true, θ is determined by the service level. In this method, the fourth service level is generally taken, referring to the definition of the service level of the highway section. At this time, the traffic flow is at the lower limit of the stable flow range, and it can also be adjusted according to the actual road conditions. When the service level is level 4, θ is 0.9, and the value is taken as the value of V / C in the highway section service level. Among them, V represents the maximum service traffic volume, and C represents the basic traffic capacity.

[0137] The introduction of service level reliability indicators can provide a decoupling method for dealing with this highly coupled and computationally intensive uncertainty problem:

[0138] With the help of service level reliability, the demand can be divided into two parts: the first part is to obtain a long-term fixed lane allocation plan within a certain time period based on the service level reliability; then, by adjusting the signal control plan for each scenario, the minimum expected total delay based on the fixed lane allocation plan can be obtained.

[0139] As a further preferred implementation, the optimization iterative calculation processing of the first-stage decoupling model and the second-stage decoupling model is performed according to the gradient descent algorithm to obtain a service level reliability index combination, including:

[0140] Initialize the service level reliability set;

[0141] Inputting the service level reliability set into the first stage decoupling model to perform lane allocation calculation processing to obtain allocated lanes;

[0142] performing expected total delay calculation processing on the second-stage decoupling model according to the allocated lanes to obtain an expected total delay difference;

[0143] The service level reliability set is updated according to the total delay expectation difference, and the process returns to the step of inputting the service level reliability set into the first-stage decoupling model for lane allocation calculation processing to obtain the lane allocation step, until the total delay expectation difference is less than a preset threshold, and the service level reliability set is determined as the service level reliability indicator combination.

[0144] In the embodiment of the present invention, Figure 4 As mentioned above, the decoupled model can solve the original problem with the help of the service level reliability index:

[0145] First, the set of service level reliability of all current entrance lanes is used as input. According to the model of the first stage, the lane allocation scheme is calculated, and then the expected total delay is obtained by solving the second stage problem under the lane allocation scheme. Then, the change of the expected total delay is observed, and the set of reliability is adjusted. By continuously iterating the above optimization process, a set of service level reliability indicator combinations that can minimize the expected total delay is obtained. The lane allocation scheme obtained based on this indicator combination is the lane allocation scheme required to minimize the expected total delay, that is, the lane allocation result of the signalized intersection.

[0146] To further simplify the calculation, in this embodiment, the reserve capacity is introduced to replace the delay in the objective function of the first stage model after decoupling. The reserve capacity describes the relationship between traffic supply and traffic demand.

[0147] It should be noted that this solution will sacrifice a certain degree of solution accuracy. If you want to ensure higher calculation accuracy, you can use the method of enumerating lane allocation schemes to replace the method of solving the first stage model after decoupling in this embodiment, but the calculation efficiency is low.

[0148] The objective function after introducing reserve capacity as a substitute can be expressed as:

[0149]

[0150] Among them, μ represents the reserve capacity. ρ is the set of service level reliability of all turns of all entrances of the actual intersection.

[0151] It should be noted that the decoupled problem needs to satisfy the following constraints in addition to the original constraints (1)-(6), (8), and (9) of the two-stage stochastic programming model:

[0152] (10) Constraint on the number of service vehicles: This means ensuring that the relationship between the number of arriving vehicles, the road capacity, and the service level reliability in the definition of service level reliability is satisfied.

[0153] (11) Reserve flow conservation constraint: that is, ensuring that the allocated flow and reserve capacity can meet traffic demand.

[0154] Among them, the constraint on the number of passing vehicles is used to ensure the smooth passage of traffic, and the reserve flow constraint is used to describe the relationship between reserve capacity, traffic demand and lane flow.

[0155] The second-stage model is consistent with the second-stage model in the original two-stage stochastic programming model.

[0156] In this embodiment, the optimal reliability index combination is solved by a gradient descent algorithm, and the steps of the algorithm include:

[0157] Step 1. Calculate the initial expected total delay D based on the given ρ0 origin , and let D = D origin , where ρ0 refers to the initial solution of the service level reliability combination ρ. i，j The value of can be obtained by bringing the solution without considering volatility into the definition of service level reliability, or by continuously bringing in different initial solutions in the experiment to find a set with higher solution efficiency and better solution effect;

[0158] Step 2. Judgement | DD origin | <∈ whether it holds true, if so, terminate the calculation, otherwise go to step 3, where ∈ represents a preset threshold value, which is a minimum value close to 0, and is 0.01 in the embodiment, and can also be adjusted according to the actual operation;

[0159] Step 3. Determine DD origin <0 is true, if true, let D=D origin , ρ0=ρ, otherwise go to step 4;

[0160] Step 4. Let ρ i，j+Δρ, determine whether the result is less than 1, if it is less than 1, go to step 5. Otherwise, go to step 6. Among them, Δρ is a small disturbance value, generally 0.05, and can also be adjusted according to actual conditions;

[0161] Step 5. Let ρ i，j -Aρ, go to step 6;

[0162] Step 6. Set the current ρ i，j Substitute into the first stage problem and calculate the lane function allocation plan. Determine whether the lane plan has changed. If not, proceed to step 4. Otherwise, proceed to step 7.

[0163] Step 7. Calculate the expected delay D(ρ) of the current lane function allocation scheme i，j ), if D(ρ i，j )-D origin ≤0, then record is D(ρi, j), otherwise it is recorded as 0;

[0164] Step 8. Determine whether j is equal to j max , if not equal to j max , then let j = j + 1, return to step 4, if it is equal to j max Go to step 9;

[0165] Step 9. Determine whether i is equal to i max , if not equal to i max , then let i=i+1, j=1, return to step 4, if it is equal to i max Go to step 10;

[0166] Step 10. Calculate the descent gradient Among them, λ and γ need to be determined by experiments;

[0167] Step 11. For all i and j, let Substitute into the calculation of expected delay D(ρ);

[0168] Step 12. If D(ρ)-D origin >0 or D(ρ)-D origin <σ, go to step 2, otherwise let Then go to step 10, where σ represents a minimum value close to 0, which is 0.01 in the embodiment and can be adjusted according to the actual situation in actual operation.

[0169] In a feasible implementation of the embodiment of the present invention, Figure 5Schematic diagram of the entrance and exit lanes of a single-point intersection in an embodiment. In the embodiment, the intersection has four forks in the east, west, south and north, and each entrance has a signal light to control the intersection of three lanes in the north-south direction and four lanes in the east-west direction. The traffic signal at the intersection has two phases, and the phases are connected by all-red moments. The vehicle arrival rate varies between different cycles and scenarios, and follows a known probability distribution. At the same time, the vehicle arrival rate in each cycle is constant. Among them, the fluctuations between scenarios can be explained as changes between days in real scenarios. In the embodiment, the minimum green light time in all directions is 10 seconds. The minimum and maximum cycles are 40 seconds and 120 seconds, respectively. For simplicity, the saturated flow rate is set to 1965pcu / h / lane. All turning radii are set to 12m. The service level level is level four, so the saturation level of all lanes is 90%, and each ρ in the initial solution ρ0 i，j It is obtained directly by substituting the solution of the benchmark solution into the definition of service level reliability. The random traffic demand follows a normal distribution, and the mean value of the random demand is Q κ As shown in Table 1 below.

[0170]

[0171] Table 1 Observed mean traffic demand (pcu / h)

[0172] To represent different random fluctuations, we set the standard deviation α of the normal distribution to three levels: 0.1Q κ , 0.3Q κ , 0.5Q κ , the larger the standard deviation, the greater the random fluctuation of traffic flow. In addition, the number of random demand scenarios used in this embodiment is 50 scenarios, and 20 cycles are calculated for each scenario.

[0173] For the sake of comparison, the lane allocation scheme that maximizes the reserve capacity without considering traffic demand fluctuations in the first stage is taken as the benchmark scheme and compared with the scheme proposed in the present invention. Under the premise of real-time signal control in the second stage, the expected total delay is obtained as shown in Table 2.

[0174]

[0175] Table 2 Results comparison

[0176] Meanwhile, the comparison of the gap results is also listed in Table 2. It can be seen that when the traffic demand is high, the traffic flow has a significant impact on the delay. Even for real-time signal control, a larger fluctuation average will lead to a larger delay. Figure 6 It shows that when α=0.3Q κ When , the lane function adjustment scheme is obtained. Figure 6As shown, although the adjustment only occurs on three lanes, the expected total delay is reduced by 7.15%, significantly reducing the average vehicle delay. As shown in the embodiment, in terms of delay performance, the lane allocation plan calculated by the proposed two-stage random programming method can reduce the delay at the intersection to a certain extent compared with the benchmark method, and the traffic operation effect is significantly improved.

[0177] and Figure 1 Corresponding to the method, an embodiment of the present invention further provides an electronic device, including a processor and a memory; the memory is used to store a program; the processor executes the program to implement the method as described above.

[0178] and Figure 1 Corresponding to the method, an embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the method as described above.

[0179] The embodiment of the present invention also discloses a computer program product or a computer program, wherein the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes Figure 1 The method shown.

[0180] In summary, the embodiments of the present invention have the following advantages: In view of the randomness of traffic flow, the present invention first constructs a two-stage random programming model for the coordinated optimization of lane allocation and signal control, wherein the first stage hopes to find a lane function allocation scheme that can minimize the sum of vehicle delays in all scenarios, and the second stage is to find the signal control parameters that minimize the average vehicle delay in a certain scenario, that is, a certain traffic demand and lane function allocation scheme. In the process of solving the model, a concept of service level reliability is introduced to decouple the problem and convert it into a problem of finding a combination of reliability indicators that minimizes the expected total delay; finally, we use a gradient descent algorithm to obtain the final solution. On the one hand, the two-stage random programming model constructed by the present invention can find a lane allocation scheme that minimizes delays and has a higher robustness of the traffic system. On the other hand, by introducing the concept of service level reliability, the present invention reduces the computational complexity of the above method and improves the solution efficiency.

[0181] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0182] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0183] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0184] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0185] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0186] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0187] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0188] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0189] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A signalized intersection lane assignment method considering random traffic flow, characterized in that: The method comprises: Get random traffic flow; Performing random programming on the random traffic flow, building a model with minimum traffic delay as the objective function, and obtaining a two-stage random programming model; Decoupling the two-stage stochastic programming model according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model; The first-stage decoupling model and the second-stage decoupling model are optimized and iterated according to a gradient descent algorithm to obtain a service level reliability indicator combination; Solving the first-stage decoupling model according to the service level reliability index combination to obtain a signalized intersection lane allocation result; The random traffic flow is subjected to random programming, and a model is constructed with the minimum delay of the traffic flow as the objective function to obtain a two-stage random programming model, including: Performing random scene division processing on the random traffic flow to obtain a random scene set; Calculating the minimum delay of traffic flow according to the random scenario set to obtain the objective function; The model is constructed according to the objective function to obtain a two-stage stochastic programming model; The calculating the minimum delay of traffic flow according to the random scene set to obtain the objective function includes: The objective function includes a first-stage objective function and a second-stage objective function; Calculating the expected minimum delay of traffic flow for all scenarios in the random scenario set to obtain the objective function of the first stage; Solving the objective function of the first stage to obtain a lane allocation scheme, and performing traffic delay calculation for any one of the random scene sets according to the lane allocation scheme to obtain the objective function of the second stage; The step of calculating the traffic delay for any one of the random scene sets according to the lane allocation scheme to obtain the second-stage objective function includes: Performing traffic cycle division processing on any one scene in the random scene set according to the lane allocation scheme to obtain a traffic cycle set; Performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay; According to the lane delay of the scenario and the vehicle flow of the scenario, the objective function of the second stage is obtained; The performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay includes: Obtaining the cycle length, traffic arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles of each cycle in the traffic cycle set; The cycle length, vehicle arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles are calculated using a lane delay calculation formula to obtain the scenario lane delay.

2. The method according to claim 1, characterized in that The two-stage stochastic programming model is decoupled according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model, including: Initialize service level reliability indicator; The first stage of the two-stage stochastic programming model is subjected to a reserve capacity calculation process by using the service level reliability index to obtain a first stage decoupling model; The traffic delay calculation process of the second stage of the two-stage stochastic programming model is performed according to the first-stage decoupling model to obtain a second-stage decoupling model.

3. The method according to claim 1, characterized in that The optimizing iterative calculation processing of the first-stage decoupling model and the second-stage decoupling model according to the gradient descent algorithm to obtain a service level reliability indicator combination includes: Initialize the service level reliability set; Inputting the service level reliability set into the first stage decoupling model to perform lane allocation calculation processing to obtain allocated lanes; performing expected total delay calculation processing on the second-stage decoupling model according to the allocated lanes to obtain an expected total delay difference; The service level reliability set is updated according to the total delay expectation difference, and the process returns to the step of inputting the service level reliability set into the first-stage decoupling model for lane allocation calculation processing to obtain the lane allocation step, until the total delay expectation difference is less than a preset threshold, and the service level reliability set is determined as the service level reliability indicator combination.

4. A signal intersection lane allocation device considering random traffic flow, characterized in that: The device comprises: The first module is used to obtain random traffic flow; The second module is used to perform random programming on the random traffic flow, build a model with the minimum delay of the traffic flow as the objective function, and obtain a two-stage random programming model; The third module is used to decouple the two-stage stochastic programming model according to the service level reliability to obtain a first-stage decoupling model and a second-stage decoupling model; The fourth module is used to optimize and iterate the first-stage decoupling model and the second-stage decoupling model according to the gradient descent algorithm to obtain a service level reliability indicator combination; A fifth module is used to solve the first-stage decoupling model according to the service level reliability index combination to obtain a signalized intersection lane allocation result; The second module is used to perform random programming processing on the random traffic flow, and to construct a model with the minimum delay of the traffic flow as the objective function to obtain a two-stage random programming model, including: Performing random scene division processing on the random traffic flow to obtain a random scene set; Calculating the minimum delay of traffic flow according to the random scenario set to obtain the objective function; The model is constructed according to the objective function to obtain a two-stage stochastic programming model; The calculating the minimum delay of traffic flow according to the random scene set to obtain the objective function includes: The objective function includes a first-stage objective function and a second-stage objective function; Calculating the expected minimum delay of traffic flow for all scenarios in the random scenario set to obtain the objective function of the first stage; Solving the objective function of the first stage to obtain a lane allocation scheme, and performing traffic delay calculation for any one of the random scene sets according to the lane allocation scheme to obtain the objective function of the second stage; The step of calculating the traffic delay for any one of the random scene sets according to the lane allocation scheme to obtain the second-stage objective function includes: Performing traffic cycle division processing on any one scene in the random scene set according to the lane allocation scheme to obtain a traffic cycle set; Performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay; According to the lane delay of the scenario and the vehicle flow of the scenario, the objective function of the second stage is obtained; The performing lane delay calculation processing on the traffic cycle set to obtain a scene lane delay includes: Obtaining the cycle length, traffic arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles of each cycle in the traffic cycle set; The cycle length, vehicle arrival rate, lane green light duration, saturation flow rate and number of remaining queued vehicles are calculated using a lane delay calculation formula to obtain the scenario lane delay.

5. An electronic device, characterized in that: The electronic device comprises a memory and a processor; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 3.

6. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images