A bus priority method based on game coordination mechanism

Abstract

The application provides a bus priority method based on a game coordination mechanism, comprising the following steps: constructing a traffic game model between a bus and other vehicles, considering a strategy space and an utility function of each bus at an intersection, and the utility function in the model can reflect a traffic efficiency of the bus; according to traffic states of the bus and the other vehicles, an optimization algorithm is used to solve the game model, and an optimal strategy of the bus is obtained; according to the optimal strategy of the bus, a traffic signal timing is dynamically adjusted, and a bus priority strategy is provided; in addition, a special situation that different phases of the intersection have buses arriving at the same time can be handled, and the bus can realize priority passing at the intersection, while reducing the influence on the other vehicles. The method of the application not only considers the priority of the bus, but also takes into account the traffic demand of the other vehicles, and realizes the target of traffic coordination.

Description

Technical Field

[0001] This invention relates to the field of intelligent traffic management, and in particular to a public transport priority method based on a game-theoretic coordination mechanism. Background Technology

[0002] In current urban traffic management, public buses, as a large-scale public transportation tool, significantly impact the smooth flow of urban traffic due to their punctuality and operational efficiency. However, the complexity and unpredictability of urban traffic, such as road conditions, traffic flow, and traffic accidents, can all lead to bus delays, making improving bus operational efficiency a crucial issue in urban traffic management. Currently, bus priority systems are implemented in many cities. These systems primarily rely on fixed signal timing schemes to improve bus priority, such as by adding dedicated bus lanes or setting bus priority signals. However, these schemes often overlook dynamic changes in traffic conditions, such as variations in traffic flow and intersection conflicts, meaning buses may still encounter delays. In recent years, some research has begun to incorporate game theory to solve traffic optimization problems, especially bus priority problems. These studies treat intersections as game scenarios, designing reasonable utility functions and policy spaces to guide buses and other vehicles in cooperative game play, thereby achieving bus priority. However, these studies typically rely on complex mathematical models and algorithms, requiring substantial computational resources, and are somewhat dependent on intersection design and vehicle behavior, limiting their application in real-world scenarios. In addition, the current solutions for dealing with the problem of buses arriving at intersections at different phases simultaneously mainly rely on empirical rules, such as fixed priority and polling. These solutions cannot be dynamically adjusted according to the actual traffic conditions and often fail to achieve optimal results. Summary of the Invention

[0003] To address the shortcomings of the existing technologies, the present invention aims to provide a bus priority method based on a game theory coordination mechanism. This method, by introducing game theory and optimization algorithms, can better solve the aforementioned problems, achieve priority for bus passage, improve bus traffic efficiency, and reduce the impact on other vehicles.

[0004] To achieve the above objectives, the technical solution of the present invention is as follows:

[0005] This invention provides a bus priority method based on a game-theoretic coordination mechanism, comprising:

[0006] Using roadside sensing devices, traffic flow information at intersections is detected, and embedded algorithms are used to detect the arrival of buses.

[0007] Using information such as perceived intersection traffic parameters, an intersection game model is established, defining the strategy space and corresponding utility function for buses and other vehicles when passing through the intersection.

[0008] Solve the game model by using an optimization algorithm based on the traffic conditions of buses and other vehicles to obtain the optimal strategy for each bus.

[0009] Based on the optimal strategy for buses, the traffic signal timing at intersections is dynamically adjusted to provide bus priority strategies.

[0010] Considering the special case where buses arrive at different phases of an intersection simultaneously, the signal timing of each phase is flexibly adjusted based on the results of the game theory model to give priority to buses while minimizing the impact on other vehicles.

[0011] Preferably, the information collection includes: using roadside sensing devices, such as intelligent high-definition cameras, radar, and radar-visual integrated machines, to collect traffic flow information at intersections;

[0012] This information includes, but is not limited to, vehicle location, speed, traffic flow, queue length in each lane, parking delays, and saturation levels;

[0013] The system detects the arrival of buses by recognizing vehicle type and license plate.

[0014] Preferably, defining the strategy space includes: the strategy space defines the set of all possible actions that each vehicle can take;

[0015] Each vehicle's action is to choose a time to pass through the intersection;

[0016] Set the passage time of all vehicles to be within a sufficiently large time window T;

[0017] Therefore, for each vehicle i, its strategy x_i should satisfy the following condition:

[0018] x_i ∈ [0, T]

[0019] Wherein, the passage time x_i of vehicle i must be within the time window [0, T];

[0020] In this strategy space, each vehicle can freely choose its travel time, as long as the time window limit is met.

[0021] Preferably, the definition of the utility function includes: the utility function is a function that measures the effectiveness of each vehicle strategy;

[0022] Buses and private vehicles have different objectives, therefore different utility functions need to be defined;

[0023] The bus utility function states that the bus's primary goal is to pass through the intersection as quickly as possible.

[0024] The utility function of a bus can be defined as the negative of its travel time, i.e.:

[0025] U_i(x_i, x_{-i}) = -x_i

[0026] Where x_i represents the travel time of bus i, and x_{-i} represents the travel time of other vehicles besides i;

[0027] Because buses want to get through as quickly as possible, the shorter the travel time, the greater the utility value.

[0028] The utility function of private vehicles includes the goals of passing through intersections as quickly as possible and avoiding time conflicts with buses.

[0029] The utility function of private vehicles can be defined as the negative of their travel time minus the penalty term for conflict with public transportation, i.e.:

[0030] U_i(x_i, x_{-i}) = -x_i - λ∑_{j∈N}φ(x_i - x_j)

[0031] Where -x_i represents the travel time of private vehicles, and -λ∑_{j∈N}φ(x_i - x_j) is the penalty for conflict with public buses;

[0032] Here, λ is a weighting parameter used to balance the travel time of private vehicles and the conflict with buses, and φ() is a penalty function. When the travel time of private vehicles conflicts with that of buses, that is, when x_i is close to x_j, the value of the φ function is larger, which imposes a greater penalty on the utility of private vehicles.

[0033] Preferably, after constructing the game model, it is necessary to select an appropriate method to achieve the optimal strategy of the utility function, including: gradient descent is an effective method to find the optimal strategy of the utility function;

[0034] It mainly includes the following five steps:

[0035] Step 1: Initialize the strategy. At the beginning, an initial strategy needs to be set.

[0036] This strategy involves the travel time of buses and other vehicles, or the timing of traffic signals.

[0037] It can be set to an initial value, such as the average travel time at the intersection;

[0038] Step 2, calculate the gradient, which requires calculating the gradient of the utility function with respect to the policy;

[0039] The gradient of the utility function U_i(s) with respect to the policy s is expressed as:

[0040] U_i(s) = [ U_i(s) / s_1, U_i(s) / s_2, ..., U_i(s) / s_n]

[0041] Where s = [s_1, s_2, ..., s_n] is the policy vector, n is the number of policies, and the policy is the travel time or signal timing of the bus and other vehicles;

[0042] Step 3: Update the policy based on gradient calculation;

[0043] Let the policy move in the opposite direction of the gradient. Because we want to maximize the utility function, the update formula for policy s is:

[0044] s_new = s_old - α U_i(s_old)

[0045] Here, α is the learning rate, which controls the step size of policy updates. s_old and s_new represent the policies before and after the update, respectively.

[0046] Step 4: Check convergence. This requires checking whether the strategy has converged.

[0047] Check if the change in the policy is less than a certain threshold;

[0048] If the change in the policy is less than this threshold, then the policy can be considered to have converged, and the iteration can be stopped.

[0049] Otherwise, we need to return to step 2 and continue iterating;

[0050] Step 5: Solve for the Nash equilibrium. When the policy converges, the Nash equilibrium has been found.

[0051] Nash equilibrium is the optimal strategy, that is, the strategy under which the utility function value is maximized;

[0052] Preferably, after solving the game model, the optimal strategy for each traffic participant, i.e. the best travel time, is obtained, including: these results need to be used to adjust the signal timing to achieve priority passage for public transport.

[0053] Signal timing adjustments mainly include the following three methods:

[0054] Method 1: Extend the green light;

[0055] Suppose that the bus arrives at the intersection at time t, and the current phase is green, denoted as G(t);

[0056] Assuming the initial green light duration is T0, and the optimal travel time for buses obtained through the game theory model is Tb, it is necessary to determine the extended green light duration ΔT.

[0057] This extended duration can be calculated using the following formula:

[0058] ΔT = max(0, Tb - T0)

[0059] If the optimal travel time Tb for the bus is greater than the initial green light duration T0, then the green light needs to be extended for a duration of Tb - T0; otherwise, the green light does not need to be extended and ΔT is 0.

[0060] Method 2: Stop red lights early;

[0061] Suppose that the bus arrives at the intersection at time t, and the current phase is red, denoted as R(t);

[0062] Assuming the remaining red light duration is Tr, and the optimal travel time for the bus obtained through the game theory model is Tb, it is necessary to determine the duration ΔT for cutting off the red light earlier.

[0063] This advance time can be calculated using the following formula:

[0064] ΔT = max(0, Tr - Tb)

[0065] If the optimal travel time Tb for the bus is less than the remaining duration Tr of the red light, then the red light needs to be turned off earlier, and the amount of time to turn off earlier is Tr - Tb; otherwise, the red light does not need to be turned off earlier, and ΔT is 0.

[0066] Method 3: Phase insertion;

[0067] For cities with dedicated bus lanes and dedicated bus phases, buses can be given priority by inserting an additional bus phase.

[0068] Let the duration of this bus phase be Tp, which can be calculated using the following formula:

[0069] Tp = Tb

[0070] Among them, the duration of the bus phase should be equal to the optimal travel time Tb of the bus;

[0071] Compared with the prior art, the beneficial technical effects of the present invention are as follows:

[0072] Compared with the prior art, the present invention has the following significant advantages:

[0073] Public transport priority: In most existing traffic signal control systems, the passage needs of all vehicles are usually treated equally. However, this invention introduces a coordination model that prioritizes public transport needs. This not only ensures the efficiency of public transport operations but also encourages more people to choose public transport as their mode of transportation, thereby helping to alleviate urban traffic congestion.

[0074] Comprehensive Optimization: This invention employs game theory to adjust traffic signal timings, rather than simply prioritizing or using fixed schedules. This approach prioritizes buses while minimizing the impact on other vehicles. This comprehensive optimization method allows the invention to improve bus efficiency while maintaining overall efficient traffic flow.

[0075] Handling of special cases: This invention also considers special cases, such as when buses arrive at an intersection from different directions simultaneously. In this situation, the model can dynamically adjust its strategy based on the actual situation, thereby avoiding potential conflicts and ensuring the passage of buses. This is something that many existing technologies cannot achieve. Attached Figure Description

[0076] Figure 1 A flowchart of a bus priority method based on a game-theoretic coordination mechanism is shown. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this invention clearer, the method proposed by this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of this invention will become clearer from the following description. It should be noted that the drawings are in a very simplified form and use non-precise proportions, only for the purpose of conveniently and clearly illustrating the embodiments of this invention. Please refer to the accompanying drawings to make the objectives, features, and advantages of this invention more apparent and understandable. It should be understood that the structures, proportions, sizes, etc., depicted in the accompanying drawings are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the implementation conditions of this invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationships, or adjustments to the size, without affecting the effects and objectives achieved by this invention, should still fall within the scope of the technical content disclosed in this invention.

[0078] This invention is mainly based on a bus priority method using a game-theoretic coordination mechanism.

[0079] like Figure 1 This invention provides a bus priority method based on a game-theoretic coordination mechanism, comprising:

[0080] Using roadside sensing devices, traffic flow information at intersections is detected, and embedded algorithms are used to detect the arrival of buses.

[0081] Using information such as perceived intersection traffic parameters, an intersection game model is established, defining the strategy space and corresponding utility function for buses and other vehicles when passing through the intersection.

[0082] Solve the game model by using an optimization algorithm based on the traffic conditions of buses and other vehicles to obtain the optimal strategy for each bus.

[0083] Based on the optimal strategy for buses, the traffic signal timing at intersections is dynamically adjusted to provide bus priority strategies.

[0084] Considering the special case where buses arrive at different phases of an intersection simultaneously, the signal timing of each phase is flexibly adjusted based on the results of the game theory model to give priority to buses while minimizing the impact on other vehicles.

[0085] S1 roadside sensing equipment senses traffic flow information at intersections. Using roadside sensing equipment, including but not limited to intelligent high-definition cameras, radar, and integrated radar-visual equipment, it senses traffic flow information at intersections, including vehicle location, speed, traffic flow volume, queue length of each lane, parking delay, saturation, etc.

[0086] S2 uses traffic flow information to establish an intersection game coordination model. It uses perceived intersection traffic parameters, intersection timing schemes, real-time intersection signal status, game relationships between different traffic phases at the intersection, and game relationships between bus traffic demand and social vehicle traffic demand to establish an intersection game coordination model.

[0087] S3 enables real-time game theory calculations. Based on real-time traffic flow information and the established game coordination model, it performs real-time game theory calculations to obtain the optimal travel strategy for each traffic participant.

[0088] The S4 signal timing adjustment method includes: extending the green light, ending the red light earlier, and inserting a phase. This is based on the results of game theory calculations. If the current phase is in a green light, the green light duration can be extended; if the current phase is in a red light, the green light can be switched to earlier; if there are dedicated bus lanes and dedicated bus phases, the bus phase can be inserted.

[0089] S5 buses are given priority passage, which has achieved the goal of giving priority passage to buses;

[0090] This invention proposes a bus priority method based on a game-theoretic coordination mechanism. This method includes the selection and use of sensing devices, the establishment of a game-theoretic coordination model, and the adjustment methods and principles for signal timing. By introducing game theory and optimization algorithms to achieve bus priority, the operating efficiency of buses can be improved, while reducing the impact on other vehicles and other public vehicles, thereby improving traffic efficiency and passenger travel convenience.

[0091] Preferably, the information collection includes: using roadside sensing devices, such as intelligent high-definition cameras, radar, and radar-visual integrated machines, to collect traffic flow information at intersections;

[0092] This information includes, but is not limited to, vehicle location, speed, traffic flow, queue length in each lane, parking delays, and saturation levels;

[0093] The system detects the arrival of buses by recognizing vehicle type and license plate.

[0094] Preferably, defining the strategy space includes: the strategy space defines the set of all possible actions that each vehicle can take;

[0095] Each vehicle's action is to choose a time to pass through the intersection;

[0096] Set the passage time of all vehicles to be within a sufficiently large time window T;

[0097] Therefore, for each vehicle i, its strategy x_i should satisfy the following condition:

[0098] x_i ∈ [0, T]

[0099] Wherein, the passage time x_i of vehicle i must be within the time window [0, T];

[0100] In this strategy space, each vehicle can freely choose its travel time, as long as the time window limit is met.

[0101] Preferably, the definition of the utility function includes: the utility function is a function that measures the effectiveness of each vehicle strategy;

[0102] Buses and private vehicles have different objectives, therefore different utility functions need to be defined;

[0103] The bus utility function states that the bus's primary goal is to pass through the intersection as quickly as possible.

[0104] The utility function of a bus can be defined as the negative of its travel time, i.e.:

[0105] U_i(x_i, x_{-i}) = -x_i

[0106] Where x_i represents the travel time of bus i, and x_{-i} represents the travel time of other vehicles besides i;

[0107] Because buses want to get through as quickly as possible, the shorter the travel time, the greater the utility value.

[0108] The utility function of private vehicles includes the goals of passing through intersections as quickly as possible and avoiding time conflicts with buses.

[0109] The utility function of private vehicles can be defined as the negative of their travel time minus the penalty term for conflict with public transportation, i.e.:

[0110] U_i(x_i, x_{-i}) = -x_i - λ∑_{j∈N}φ(x_i - x_j)

[0111] Where -x_i represents the travel time of private vehicles, and -λ∑_{j∈N}φ(x_i - x_j) is the penalty for conflict with public buses;

[0112] Here, λ is a weighting parameter used to balance the travel time of private vehicles and the conflict with buses, and φ() is a penalty function. When the travel time of private vehicles conflicts with that of buses, that is, when x_i is close to x_j, the value of the φ function is larger, which imposes a greater penalty on the utility of private vehicles.

[0113] Preferably, after constructing the game model, it is necessary to select an appropriate method to achieve the optimal strategy of the utility function, including: gradient descent is an effective method to find the optimal strategy of the utility function;

[0114] It mainly includes the following five steps:

[0115] Step 1: Initialize the strategy. At the beginning, an initial strategy needs to be set.

[0116] This strategy involves the travel time of buses and other vehicles, or the timing of traffic signals.

[0117] It can be set to an initial value, such as the average travel time at the intersection;

[0118] Step 2, calculate the gradient, which requires calculating the gradient of the utility function with respect to the policy;

[0119] The gradient of the utility function U_i(s) with respect to the policy s is expressed as:

[0120] U_i(s) = [ U_i(s) / s_1, U_i(s) / s_2, ..., U_i(s) / s_n]

[0121] Where s = [s_1, s_2, ..., s_n] is the policy vector, n is the number of policies, and the policy is the travel time or signal timing of the bus and other vehicles;

[0122] Step 3: Update the policy based on gradient calculation;

[0123] Let the policy move in the opposite direction of the gradient. Because we want to maximize the utility function, the update formula for policy s is:

[0124] s_new = s_old - α U_i(s_old)

[0125] Here, α is the learning rate, which controls the step size of policy updates. s_old and s_new represent the policies before and after the update, respectively.

[0126] Step 4: Check convergence. This requires checking whether the strategy has converged.

[0127] Check if the change in the policy is less than a certain threshold;

[0128] If the change in the policy is less than this threshold, then the policy can be considered to have converged, and the iteration can be stopped.

[0129] Otherwise, we need to return to step 2 and continue iterating;

[0130] Step 5: Solve for the Nash equilibrium. When the policy converges, the Nash equilibrium has been found.

[0131] Nash equilibrium is the optimal strategy, that is, the strategy under which the utility function value is maximized;

[0132] Preferably, after solving the game model, the optimal strategy for each traffic participant, i.e. the best travel time, is obtained, including: these results need to be used to adjust the signal timing to achieve priority passage for public transport.

[0133] Signal timing adjustments mainly include the following three methods:

[0134] Method 1: Extend the green light;

[0135] Suppose that the bus arrives at the intersection at time t, and the current phase is green, denoted as G(t);

[0136] Assuming the initial green light duration is T0, and the optimal travel time for buses obtained through the game theory model is Tb, it is necessary to determine the extended green light duration ΔT.

[0137] This extended duration can be calculated using the following formula:

[0138] ΔT = max(0, Tb - T0)

[0139] If the optimal travel time Tb for the bus is greater than the initial green light duration T0, then the green light needs to be extended for a duration of Tb - T0; otherwise, the green light does not need to be extended and ΔT is 0.

[0140] Method 2: Stop red lights early;

[0141] Suppose that the bus arrives at the intersection at time t, and the current phase is red, denoted as R(t);

[0142] Assuming the remaining red light duration is Tr, and the optimal travel time for the bus obtained through the game theory model is Tb, it is necessary to determine the duration ΔT for cutting off the red light earlier.

[0143] This advance time can be calculated using the following formula:

[0144] ΔT = max(0, Tr - Tb)

[0145] If the optimal travel time Tb for the bus is less than the remaining duration Tr of the red light, then the red light needs to be turned off earlier, and the amount of time to turn off earlier is Tr - Tb; otherwise, the red light does not need to be turned off earlier, and ΔT is 0.

[0146] Method 3: Phase insertion;

[0147] For cities with dedicated bus lanes and dedicated bus phases, buses can be given priority by inserting an additional bus phase.

[0148] Let the duration of this bus phase be Tp, which can be calculated using the following formula:

[0149] Tp = Tb

[0150] Among them, the duration of the bus phase should be equal to the optimal travel time Tb of the bus;

[0151] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0152] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A bus priority method based on a game coordination mechanism, characterized in that, Comprise: Using road end sensing devices, sensing intersection traffic flow information, and combining embedded algorithm to perceive the arrival of buses; Using the sensed intersection traffic parameters and other information, establishing an intersection game model, setting the strategy space of buses and other vehicles at the intersection, and the corresponding utility function; Solving the game model, according to the traffic state of buses and other vehicles, using optimization algorithm to solve the game model, and obtaining the optimal strategy of each bus; Build a game model, which includes two types of participants: buses and social vehicles; In order to simulate the decision-making process of these two types of vehicles, it is necessary to define the strategy space and utility function; Strategy space definition: the strategy space of each participating vehicle is to choose the time to pass through the intersection; Assume that the passing time of all vehicles is within a large enough time window T; Therefore, for each vehicle i, its strategy x_i should satisfy the following formula: 0 ≤ x_i ≤ T Utility function definition: the utility function of buses and social vehicles should reflect the goal; For buses, the goal is to pass through the intersection as soon as possible, so the utility function can be defined as: U_i(x_i, x_{-i}) = -x_i Where x_{-i} represents the strategy of other vehicles; For social vehicles, they also want to pass through as soon as possible, but at the same time need to avoid time conflicts with buses, so the utility function can be defined as: U_i(x_i, x_{-i}) = -x_i - λ∑_{j∈N}φ(x_i - x_j), where N is a set of positive integers Here λ is a weight parameter to balance the passing time of social vehicles and the conflict with buses, and φ() is a penalty function that penalizes social vehicles when their passing time conflicts with buses, i.e. x_i is close to x_j, φ() has a large value, which greatly penalizes the utility of social vehicles; Where x_j represents the jth strategy; According to the optimal strategy of buses, dynamically adjust the traffic signal timing of the intersection, and provide bus priority strategy; Solve the Nash equilibrium of this game, that is, in this equilibrium state, no one can improve their utility by changing their strategy alone; In order to find this Nash equilibrium, the gradient descent optimization algorithm can be used; Gradient descent is an iterative algorithm for solving optimal solutions, its basic idea is to use the gradient direction of the objective function to update the parameters; First, randomly initialize the strategy of all vehicles: x = {x_i}_{i∈N}, where N is a set of positive integers Then in each step, for each vehicle i, update its strategy as follows: x_i' = x_i + α U_i / x_i where a is the learning rate, which determines the step size of the parameter update, U_i / x_i is the gradient of the utility function U_i with respect to the policy x_i; This process is repeated until the strategies x of all vehicles converge, i.e. the change in x over two consecutive iterations is less than a predetermined threshold ε, and a Nash equilibrium strategy x is obtained ; Consider the special case of buses arriving at the intersection at different phases, according to the results of the game model, flexibly adjust the signal timing of each phase, realize the priority of buses, and at the same time minimize the impact on other vehicles.

2. The bus priority method based on game coordination mechanism according to claim 1, characterized in that, Collect information, use road end sensing devices such as intelligent high-definition cameras, radars, radar and video integrated machines, etc. to collect intersection traffic flow information; These information includes but not limited to vehicle position, speed, traffic flow, queue length of each lane, parking delay and saturation; Through vehicle type recognition and license plate recognition, the arrival of buses is perceived.

3. The bus priority method based on game coordination mechanism according to claim 1, characterized in that, According to the game results, the optimal transit time of each vehicle is known, and then the signal timing can be adjusted according to the following principles: Green light extension: If the bus arrives at the intersection when the current phase is in green light, the green light duration of the current phase can be considered to be extended to ensure that the bus can pass through the intersection; Red light early break: If the bus arrives at the intersection when the current phase is in red light, the red light can be considered to be ended early to turn green light, so that the bus can pass through the intersection early; Phase insertion: If there is a bus lane and a bus phase, a bus phase can be inserted to make the bus have priority; For special cases of intersections, such as multiple buses arriving in multiple directions at the same time, they can be treated as multiple independent game problems, and then the game results of each direction are combined to adjust the signal timing.

Images