Time prediction-based bus lane priority control method

Abstract

The present application belongs to the technical field of traffic signal control, and discloses a bus priority driving control method based on time prediction without special lane, comprising the following steps: S1, obtaining relevant parameters and predicting the time for a bus to reach the start point of a speed guidance area; S2, according to the remaining green light time of a downstream intersection, if the bus cannot pass the downstream intersection without stopping, a bus acceleration guidance strategy and / or a green light extension strategy are executed; S3, according to the remaining red light time of the downstream intersection, if the bus cannot wait for the next signal cycle without stopping, a bus deceleration guidance strategy and / or a red light early break strategy are executed. The present application designs bus priority driving decision mechanisms in various scenarios, providing a reference for improving the service level of urban public transportation.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of traffic signal control, and particularly relates to a bus vehicle priority driving control method based on time prediction without a special lane. BACKGROUND

[0002] Bus signal priority can effectively improve the efficiency and reliability of the bus operation system, and has important significance for relieving urban traffic congestion. Traditional bus signal priority control is divided into passive priority, active priority and adaptive priority. There are relatively mature theoretical results at home and abroad to realize the priority of bus. According to the triggering of traffic events, Dong Yupu et al. designed a bus signal priority control strategy based on phase priority rules in a double-ring phase structure. Xu Hongfeng et al. studied the control method based on logic rules. However, most of the above researches predict the time of bus vehicles arriving at the intersection, thereby adjusting the traffic signal light state, ignoring the influence of the fluctuation of intersection social vehicle arrival and the uncertainty of queue length on the priority control effect. But the above strategy cannot completely obtain the real-time information of the bus vehicles on the road section, so it cannot effectively respond to different traffic conditions and bus vehicle operating states.

[0003] In recent years, with the development of Internet of Vehicles technology, information can be transmitted in real time and bidirectionally between vehicles and controllers, making traffic control from passive response to traffic flow to active guidance, providing more strategies to meet the needs of bus priority. Therefore, more and more scholars begin to study the bus signal priority control in the artificial-network hybrid environment, such as combining speed guidance, station control and other strategies with traditional bus priority. Ma Wanjing et al. designed the coordination optimization rules of bus vehicle running speed and signal priority control strategy with the goal of optimizing the bus running state. Zheng Chen et al. divided the arrival time of bus vehicles according to whether there is a station at the intersection, and adopted different speed induction strategies for different intervals. Oushiqi et al. established a real-time bus priority control method to solve the problem that bus priority may damage the coordination of trunk lines and affect the driving efficiency of social vehicles, realizing the integration of bus vehicle speed induction-station control and signal optimization control. Wang Baojie et al. applied Kalman to predict the bus vehicle travel time in real time, and combined with the station time and background signal timing to guide the speed of BRT vehicles, so that the vehicles can arrive at the intersection during the green light as much as possible. However, the above researches are all based on the premise of bus lane, ignoring the factor of social vehicle queue when there is no bus lane.

[0004] At present, the proportion of bus lane in urban area is not high in China, and most of the time, social vehicles and buses are mixed on the road. Considering the interference of social vehicles to buses, Yang et al. studied the release sequence optimization of buses and social vehicles at the entrance of intersection without bus lane, which can reduce the passenger delay. Wu et al. established a bus stop model for the bus arriving in advance and a speed guidance model for the bus arriving late, and introduced a signal priority control model when the speed guidance model is invalid. In order to avoid the influence of traditional bus signal priority on the coordination control of trunk, Cai Yaping et al. proposed an integrated scheme of bus priority variable speed guidance and multi-intersection signal timing optimization, and established a bus priority model of speed guidance under real-time trunk coordination control, which can greatly reduce the travel time of buses. Zeng et al. established a path-based bus signal priority model (R-TSP) and its localized model (L-TSP), analyzed and discussed the formula variants of the two models, and the R-TSP model greatly reduces the bus delay and improves the punctuality rate. Hu et al. proposed an intelligent bus signal priority logic optimization method (TSPCV-C) based on passenger delay to minimize the passenger delay, to ensure that the method has benefits in trunk. However, the above researches mainly adjust the running state or signal state of the bus after detecting the bus, and do not consider the fluctuation of the queue length during the bus driving from the guidance area to the intersection, which may lead to the decline of bus priority effect or even negative optimization in actual application. SUMMARY

[0005] The present application provides a bus priority driving control method without bus lane based on time prediction, which designs bus priority driving decision mechanism in various scenarios, and provides a reference for improving the service level of urban public transportation.

[0006] In order to achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0007] The present application provides a bus priority driving control method without bus lane based on time prediction, which includes the following steps:

[0008] S1, in the vehicle-network environment, the driving speed of the bus, the real-time distance from the bus to the stop line of the downstream intersection, the signal state of the downstream intersection when the bus reaches the starting point of the speed guidance area, and the vehicle passing situation of the downstream intersection are obtained, and the extended Kalman filter model is used to predict the time when the bus reaches the starting point of the speed guidance area;

[0009] S2, if the signal state of the downstream intersection is green when the bus reaches the start of the vehicle speed guiding area, judging whether the bus can pass the downstream intersection without stopping according to the remaining green time of the downstream intersection, if yes, maintaining the original driving speed and the original signal timing; if no, executing the bus acceleration guiding strategy and / or the green light extension strategy, and solving the minimum sum of the travel delay time and the signal delay time of each vehicle at the downstream intersection by using the genetic algorithm;

[0010] S3, if the signal state of the downstream intersection is red when the bus reaches the start of the vehicle speed guiding area, judging whether the bus can reach the downstream intersection without stopping and waiting for the next green light according to the remaining red time of the downstream intersection, if yes, maintaining the original driving speed and the original signal timing; if no, executing the bus deceleration guiding strategy and / or the red light early break strategy, and solving the minimum sum of the travel delay time and the signal delay time of each vehicle at the downstream intersection by using the genetic algorithm.

[0011] Further, the vehicle passing situation at the downstream intersection includes the vehicle flow of each passing direction at the intersection and the queue length of the social vehicles when the signal state of the downstream intersection is red; the length of the vehicle speed guiding area is the distance from the stop line of the downstream intersection to the upstream 150-200 m of the stop line.

[0012] Further, the execution of the bus acceleration guiding strategy and / or the green light extension strategy in step S2 is specifically divided into two cases:

[0013] 1) only the bus acceleration guiding strategy can make the bus pass the downstream intersection without stopping, in which case the guiding speed of the bus is calculated according to the following formula:

[0014]

[0015] Wherein: T is the time when the bus reaches the start of the vehicle speed guiding area, in seconds; L is the distance from the start of the vehicle speed guiding area to the stop line of the downstream intersection, in meters; α and β are weighting coefficients; V b V is the average speed of the bus, in km / h, and V min ≤ V b ≤ V max , V max V is the maximum speed allowed for the bus, V min V is the minimum speed of the bus; a a a is the acceleration of the bus, in m / s 2 , and is the minimum acceleration of the bus vehicle, is the maximum acceleration of the bus vehicle;

[0016] 2) When the bus vehicle acceleration guiding strategy and the green light extension strategy are executed simultaneously, the bus vehicle can pass the downstream intersection smoothly without stopping, and the guiding speed calculation formula of the bus vehicle is as follows:

[0017] V i-two = V max ;

[0018] At this time, the calculation formula of the green light extension time is as follows:

[0019]

[0020] Wherein: Δg 1j is the remaining green light time of the bus phase in the jth signal cycle, and the unit is s; and is the maximum green light extension time, and the unit is s; C is the signal cycle length of the downstream intersection, and the unit is s; λ i is the minimum green ratio of the ith phase; G i is the minimum green light time of the ith phase, and the unit is s; l is the green light loss time, and the unit is s; q is the social vehicle arrival rate of the downstream intersection, and S is the social vehicle saturation flow rate of the downstream intersection.

[0021] Further, when only the bus vehicle acceleration guiding strategy is executed, the travel delay of the bus vehicle in acceleration is smaller than that of the bus vehicle in normal speed, and the acceleration process of the bus vehicle has smaller influence on the delay of other social vehicles, so the travel delay time of the social vehicles caused by the acceleration of the bus vehicle can be ignored, and the present application only calculates the travel delay time of the bus vehicle caused by the acceleration of the bus vehicle, and the calculation formula is as follows:

[0022] ΔD b = L / (V b -V i-one ).

[0023] Further, when the bus vehicle acceleration guiding strategy and the green light extension strategy are executed simultaneously, the delay of the bus vehicle in the reduced traffic phase is equal to the time that the bus needs to wait at the intersection before the signal optimization, so the calculation formula of the signal delay time of the bus vehicle in the priority phase caused by the green light extension is as follows:

[0024]

[0025] Wherein: r ijfor the red light time of the non-priority phase.

[0026] In addition, the bus vehicle priority phase also provides additional travel time for social vehicles, so that the social vehicles in the priority phase can pass without queuing for the next signal cycle. Therefore, the calculation formula of the signal delay time of the social vehicles in the priority phase caused by the green light extension is as follows:

[0027]

[0028] wherein r 1j is the red light time of the priority phase; q 1j is the arrival rate of the social vehicles in the priority phase; S 1j is the saturation flow rate of the social vehicles in the priority phase;

[0029] After the green light extension strategy is adopted, the social vehicles arriving from the non-priority phase need to wait for g extent time to pass the downstream intersection, and the calculation formula of the signal delay time of the social vehicles in the non-priority phase caused by the green light extension is as follows:

[0030]

[0031] wherein qij is the arrival rate of the social vehicles in the non-priority phase; S ij represents the saturation flow rate of the social vehicles in the non-priority phase.

[0032] According to the above formula, when the bus vehicle acceleration guidance strategy is executed and / or the green light extension strategy is executed, the calculation model of the sum of the travel delay time and the signal delay time of each vehicle at the downstream intersection is as follows:

[0033]

[0034] wherein ΔD represents the reduced delay of the bus vehicles after optimization, Occ b represents the average occupancy rate of the bus vehicles; ΔD c represents the reduced delay of the social vehicles after optimization, Occ c represents the average occupancy rate of the social vehicles; J is the total number of signal phases of the intersection; is the number of bus vehicles passing through in the optimization period; is the total number of social vehicles passing through in the optimization period.

[0035] Further, the execution of the bus vehicle deceleration guidance strategy and / or the red light early termination strategy in step S3 is specifically divided into two cases:

[0036] 1) Only the bus vehicle deceleration guidance strategy can make the start wave of the next signal cycle green light to the bus vehicle to avoid parking and waiting, at this time the bus vehicle guidance speed calculation formula is as follows:

[0037]

[0038] Wherein: is the time of bus vehicle reaching the start point of speed guidance area, unit: s; γ, η are weight coefficients; L is the distance from the start point of speed guidance area to the stop line of downstream intersection, unit: m; V b is the average speed of bus vehicle, unit: km / h, V min is the minimum speed of bus vehicle; a d is the deceleration of bus vehicle, unit: m / s 2 , is the minimum deceleration of bus vehicle, is the maximum deceleration of bus vehicle; X (i,j) represents the vehicle queue length of downstream intersection in the i phase of the j signal cycle, unit: m, q is the arrival rate of social vehicle of downstream intersection; L v is the average length of social vehicle of downstream intersection, is the red light start time of i phase in the j signal cycle; T m is the start time of tail social vehicle S is the saturation flow rate of social vehicle of downstream intersection, is the green light start time of i phase in the j+1 cycle;

[0039] 2) Only the bus vehicle deceleration guidance strategy and the red light early break strategy can make the start wave of the next signal cycle green light to the bus vehicle to avoid parking and waiting, at this time the bus vehicle guidance speed calculation formula is as follows:

[0040] V i-four = V min ;

[0041] At this time, the calculation formula of red light early break time is as follows:

[0042]

[0043] Wherein: is the red light end time of i phase in the j signal cycle; T a is the time of bus vehicle driving to the tail at the guidance speed, and is the maximum red light early break time; G (i-1,j)T is the original green time of a phase in the direction of bus travel; (i-1,j) T is the time required for the queue of vehicles in the direction of bus travel to dissipate completely; 1j q is the red time of the priority phase; 1j S is the arrival rate of social vehicles in the priority phase; 1j S is the saturation flow rate of social vehicles in the priority phase.

[0044] Further, when only the bus vehicle deceleration guidance strategy is implemented, the bus vehicle is in a deceleration driving state, and the travel delay of the bus vehicle is greater than that of a bus vehicle driving at a normal speed. At this time, the travel delay of the bus vehicle is divided into two parts: one is the delay from driving in the vehicle speed guidance area to the end of the queue; and the other is the delay from following the queue vehicles from the end of the queue to the intersection. Therefore, the calculation formula of the travel delay time of the bus vehicle caused by the deceleration driving of the bus vehicle is as follows:

[0045]

[0046] Further, when the bus vehicle deceleration guidance strategy and the red light early break strategy are implemented at the same time, since the delay of the bus vehicle reduced after the red light early break is equal to the sum of the waiting time for the red light before signal timing adjustment and the time for the start wave to be transmitted to the end of the queue when the green light is on, the calculation formula of the signal delay time of the bus vehicle in the priority phase caused by the red light early break is as follows:

[0047]

[0048] Since the red light early break is implemented, the social vehicles consistent with the direction of bus travel can also drive away from the intersection in advance. Therefore, the calculation formula of the signal delay time of the social vehicles in the priority phase caused by the red light early break is as follows:

[0049]

[0050] In addition, after the red light early break is implemented, the green light time of the non-priority phase is compressed, and the vehicles that can drive away from the intersection before optimization may need to wait for the green light of the next cycle to pass, resulting in an increase in the time for the social vehicles in the non-priority phase to queue and wait at the intersection. Therefore, the calculation formula of the signal delay time of the social vehicles in the non-priority phase caused by the red light early break is as follows:

[0051]

[0052] q is the red time of the non-priority phase; ij S is the arrival rate of social vehicles in the non-priority phase; ij S is the saturation flow rate of social vehicles in the non-priority phase.

[0053] From the above, the calculation model of the sum of the travel delay time and the signal delay time of each vehicle at the downstream intersection when the bus vehicle deceleration guidance strategy is executed and / or the red light early break strategy is executed is as follows:

[0054]

[0055] wherein ΔD' represents the reduced delay of the bus vehicle after optimization, O'cc b represents the average occupancy rate of the bus vehicle; ΔD' c represents the reduced delay of the social vehicle after optimization, Occ c represents the average occupancy rate of the social vehicle; J is the total number of signal phases of the intersection; is the number of buses passing through in the optimization period; is the total number of social vehicles passing through in the optimization period.

[0056] Compared with the prior art, the present application has the beneficial effects that:

[0057] The traditional bus arrival time depends on the detection information and the road state ahead, which seriously restricts the effect of the priority. The bus vehicle priority driving control method based on time prediction of the present application is based on the bidirectional real-time transmission of information under the vehicle networking environment, adopts an extended Kalman filter model to predict the time of the bus vehicle arriving at the speed guidance area, further compares the time of the bus vehicle arriving at the starting point of the speed guidance area with the signal remaining time of the downstream intersection at this time, and considers the queuing phenomenon of the social vehicles at the intersection, so as to design a bus vehicle priority driving decision mechanism (acceleration, acceleration + green light extension, deceleration, deceleration + red light early break) in four scenarios to meet the priority demand of the bus with different arrival times, and provide a reference for improving the service level of urban public transportation. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 is a schematic diagram of the intersection of the present application.

[0059] Figure 2 is a schematic diagram of the scenario of only executing the bus vehicle acceleration guidance strategy of the present application.

[0060] Figure 3 is a schematic diagram of the scenario of simultaneously executing the bus vehicle acceleration guidance strategy and the green light extension strategy of the present application.

[0061] Figure 4 is an analysis diagram of the signal delay time of the social vehicle in the priority phase caused by the green light extension of the present application.

[0062] Figure 5The analysis schematic diagram of signal delay time of social vehicles in non-priority phase caused by green light extension of the present application.

[0063] Figure 6 The scene schematic diagram of the present application only executing the bus vehicle deceleration guidance strategy.

[0064] Figure 7 The scene schematic diagram of the present application simultaneously executing the bus vehicle deceleration guidance strategy and executing the red light early break strategy.

[0065] Figure 8 The analysis schematic diagram of signal delay time of social vehicles in priority phase caused by red light early break of the present application.

[0066] Figure 9 The analysis schematic diagram of signal delay time of social vehicles in non-priority phase caused by red light early break of the present application.

[0067] Figure 10 The flow chart of the present application adopting genetic algorithm to solve the minimum value of the sum of travel delay time and signal delay time of each vehicle of downstream intersection.

[0068] Figure 11 The simulation schematic diagram of the application example of the present application.

[0069] Figure 12 The intersection signal timing scheme of the application example of the present application.

[0070] Figure 13 The comparison chart of the average number of stop of bus vehicles before and after traffic volume optimization in different time periods of the application example of the present application.

[0071] Figure 14 The comparison chart of the average delay time of bus vehicles before and after traffic volume optimization in different time periods of the application example of the present application. DETAILED DESCRIPTION

[0072] The following examples are intended to illustrate the present application but not to limit the scope of protection of the present application. If not specifically indicated, the technical means used in the examples are the conventional means known to those skilled in the art. The test methods in the following examples are the conventional methods, if not specifically indicated.

[0073] Example 1

[0074] The vehicle-network-intersection relied on by the present application is as follows Figure 1As shown in the figure, and the following assumptions are made: (1) the study area is a single grid intersection, and there is no bus vehicle station near the upstream of the intersection. (2) The vehicle strictly obeys the speed guiding strategy after entering the speed guiding area. (3) The running state (speed) of the vehicle and the length of the vehicle from the intersection can be detected in real time, and the bus vehicle can communicate with the signal controller in both directions. (4) The interference of pedestrians and non-motor vehicles is not considered. (5) No bus lane is set.

[0075] When the bus vehicle reaches the grid intersection, it will generally encounter the following situations: (1) uniform speed through the intersection. When the bus vehicle enters the speed guiding area, the social vehicle queue has completely disappeared. (2) deceleration through the intersection. When the bus vehicle enters the speed guiding area, the signal light is at the beginning of the green light or the end of the red light, and the social vehicle queue at the intersection has not completely disappeared, and the vehicle needs to decelerate to pass through the intersection. (3) accelerate through the intersection. When the bus vehicle enters the guiding area, the signal light is at the end of the green light, and the vehicle needs to increase the speed to pass through the intersection before the end of the green light of this period. (4) stop and wait for the next cycle green light to pass through the intersection.

[0076] Based on this, the application is a kind of bus vehicle priority driving control method based on time prediction without special lane, comprising the following steps:

[0077] S1, in the vehicle-grid environment, the driving speed of the bus vehicle, the real-time distance of the bus vehicle to the parking line of the downstream intersection, the signal state of the downstream intersection when the bus vehicle reaches the start point of the speed guiding area, and the vehicle passing condition of the downstream intersection are obtained, and the extended Kalman filter model is used to predict the time when the bus vehicle reaches the start point of the speed guiding area.

[0078] The traditional Kalman filter (KF) is generally based on the uniform motion of the vehicle or the simple acceleration and deceleration to describe the dynamics and observation equation of the vehicle, so the model can be regarded as a linear model. Under the condition of no bus lane, that is, the bus vehicle and the social vehicle are mixed, the interaction between the two, the road congestion and the signal timing factors need to be considered, which will lead to the nonlinearity of the dynamics and observation equation, so the extended Kalman filter (EKF) is used to solve the non-linear state measurement of the bus vehicle, and more accurate state estimation and prediction results are provided.

[0079] When the bus vehicle starts from the upstream intersection, the EKF is used to predict the time when the bus vehicle reaches the starting point of the speed guidance area and the signal state (green light, red light, yellow light) at that time, so as to give the constructed model more reaction time and more accurately implement speed guidance and signal optimization on the bus vehicle, so that it can smoothly pass through the intersection without stopping. The driving process of the bus vehicle from the upstream intersection to the guidance area point can be regarded as the driving state of the vehicle. The state parameters mainly include the length of the vehicle from the intersection and the vehicle speed. According to this, the nonlinear dynamics equation and the observation equation of the bus vehicle are as follows:

[0080] x k+1 =f(x k ,u k )+w k ,

[0081] z k =h(x k ,u k )+v k ,

[0082] Wherein: x k+1 represents the nonlinear dynamics equation; f(x k ,u k ) represents the nonlinear state transition function; w k represents the Gaussian white noise; z k represents the nonlinear observation equation; h(x k ,u k ) represents the nonlinear observation function; v k represents the observation noise vector; and the covariance is Q or R.

[0083] In order to further improve the prediction accuracy of the model, the following state variables and control variables are defined:

[0084] x k =(l k ,v k ) T ,

[0085] u k =(Q k ,s k ) T ,

[0086] Wherein: l k represents the position information of the bus vehicle; v k represents the driving speed of the bus vehicle; Q k represents the congestion condition of the road section; and s k represents the intersection signal timing information.

[0087] The Taylor expansion of the nonlinear dynamic equation and the observation equation of the bus vehicle is carried out and the highest term is discarded:

[0088]

[0089]

[0090] Definition:

[0091]

[0092]

[0093]

[0094] By Taylor expansion of the nonlinear function, the above nonlinear equation is converted into a linear equation, the initial value is assigned and the recursive estimation of the extended Kalman filter is carried out, and the model input is u k-1 , the recursive process is as follows:

[0095]

[0096] p k / k-1 =A k-1 p k-1 / k-1 A k-1 T +Q k-1 ,

[0097] K k =o k / k-1 H k T [H k p k / k-1 H k T +R k-1 ] -1 ,

[0098] p k / k =[I-K k H k ]p k / k-1 ,

[0099] Then the formula for predicting the time of the bus vehicle reaching the starting point of the speed guide area by the extended Kalman filter model is as follows:

[0100]

[0101] S2, if the signal state of the downstream intersection is green when the public transport vehicle reaches the start point of the vehicle speed guiding area, judging whether the public transport vehicle can pass the downstream intersection without stopping according to the remaining green time of the downstream intersection, if yes, maintaining the original driving speed and original signal timing, if no, executing the public transport vehicle acceleration guiding strategy and / or executing the green light extension strategy, and using a genetic algorithm to solve the minimum sum of travel delay time and signal delay time of each vehicle at the downstream intersection (i.e. minimizing the total person delay).

[0102] The present application specifically divides the execution of the public transport vehicle acceleration guiding strategy and / or the green light extension strategy into two cases:

[0103] 1) only the public transport vehicle acceleration guiding strategy can make the public transport vehicle pass the downstream intersection without stopping (i.e. as long as the public transport vehicle accelerates, it can pass the downstream intersection without stopping before the green light of the downstream intersection ends, as shown in FIG. 2), at this time, the guiding speed calculation formula of the public transport vehicle is as follows: Figure 2

[0104]

[0105] Among them: is the time when the public transport vehicle reaches the start point of the vehicle speed guiding area, unit: s; is the green light end time of the i phase in the j signal cycle, unit: s; L is the distance from the start point of the vehicle speed guiding area to the stop line of the downstream intersection, unit: m; α and β are weighting coefficients; V b is the average speed of the public transport vehicle, unit: km / h, and V min ≤ V b ≤ V max , V max is the maximum speed allowed by the public transport vehicle, V min is the minimum speed of the public transport vehicle; a a is the acceleration of the public transport vehicle, unit: m / s 2 , and is the minimum acceleration of the public transport vehicle, is the maximum acceleration of the public transport vehicle.

[0106] When only the public transport vehicle acceleration guiding strategy is executed, the travel delay of the public transport vehicle in acceleration is smaller than that of the public transport vehicle in normal speed, and the acceleration process of the public transport vehicle has less impact on the delay of other social vehicles, so the travel delay time of the social vehicles caused by the acceleration of the public transport vehicle can be ignored, the present application only calculates the travel delay time of the public transport vehicle caused by the acceleration of the public transport vehicle, and the calculation formula is as follows:

[0107] ΔD b ​= L / (V b -V i-one )。

[0108] 2) When the bus acceleration guidance strategy and the green light extension strategy are executed simultaneously, the bus can pass the downstream intersection smoothly without stopping (i.e., only accelerating the bus to the maximum speed allowed for the bus cannot make the bus pass the downstream intersection smoothly without stopping, and the green light must be extended to make the bus pass the downstream intersection smoothly without stopping, as shown in FIG. 2), and the guidance speed of the bus is calculated according to the following formula: Figure 3

[0109] V i-two = V max .

[0110] The calculation formula of the green light extension time is as follows:

[0111]

[0112] wherein Δg 1j is the remaining green light time of the bus phase in the jth signal cycle, in seconds; and is the maximum green light extension time, in seconds; C is the length of the signal cycle of the downstream intersection, in seconds; λ i is the minimum green ratio of the ith phase; G i is the minimum green light time of the ith phase, in seconds; l is the green light loss time, in seconds; and q is the arrival rate of the social vehicles at the downstream intersection, and S is the saturation flow rate of the social vehicles at the downstream intersection.

[0113] It is worth noting that the green light extension will result in a decrease in the green light time of the remaining phases without changing the length of the signal cycle, and therefore, in order to avoid excessive saturation (≥ 0.9) of the traffic of the remaining phases at the intersection, the maximum green light extension time needs to be specified.

[0114] When the bus acceleration guidance strategy and the green light extension strategy are executed simultaneously, the delay of the bus in the reduced phase is equal to the time that the bus needs to wait at the intersection before the signal optimization, and therefore, the calculation formula of the signal delay time of the bus in the priority phase due to the green light extension is as follows:

[0115]

[0116] wherein r ij is the red light time of the non-priority phase.

[0117] ​Furthermore, the priority phase for buses also grants extra passage time to other vehicles, allowing them to pass through without queuing for the next signal cycle. Figure 4 As shown, the formula for calculating the signal delay time of social vehicles in the priority phase due to the extension of the green light is as follows:

[0118]

[0119] Where: r 1j Red light duration for priority phase; q 1j Priority phase social vehicle arrival rate; S 1j The priority phase is the saturation flow rate of social vehicles.

[0120] After the green light extension strategy is adopted, vehicles arriving from directions other than the priority phase need to wait additionally. extent Time is needed to pass through the downstream intersection, such as Figure 5 As shown, the formula for calculating the signal delay time of social vehicles in non-priority phases due to the extension of green lights is as follows:

[0121]

[0122] Where: q ij For non-priority phase social vehicle arrival rate; S ij This indicates the saturation flow rate of social vehicles in the non-priority phase.

[0123] As shown in the above formula, when implementing the bus acceleration guidance strategy and / or the green light extension strategy, the calculation model for the sum of the travel delay time and signal delay time for each vehicle at the downstream intersection is as follows:

[0124]

[0125] Where: ΔD represents the reduction in delays caused by the optimized bus service. Occ b ΔD represents the average occupancy rate of public transport vehicles. c This indicates the delay caused by the reduction in social vehicles after optimization. Occ c This represents the average occupancy rate of social vehicles; J is the total number of signal phases at the intersection. To optimize the number of buses passing through within a time period; To optimize the total number of social vehicles passing through within a time period.

[0126] S3, if the signal state of the downstream intersection is red when the bus vehicle reaches the start point of the speed guidance area, according to the remaining time of the red light of the downstream intersection, it is judged whether the bus vehicle can reach the downstream intersection without stopping and queuing to wait for the green light of the next signal period at the current speed, if yes, the original driving speed and the original signal timing are maintained, if not, the bus vehicle deceleration guidance strategy and / or the red light early breaking strategy are executed, and the genetic algorithm is used to solve the minimum sum of the travel delay time and the signal delay time of each vehicle at the downstream intersection (i.e. the minimum total human delay minimization).

[0127] The bus vehicle deceleration guidance strategy and / or the red light early breaking strategy executed in step S3 are specifically divided into two cases:

[0128] 1) only the bus vehicle deceleration guidance strategy can make the start wave of the green light of the next signal period reach the bus vehicle to avoid stopping and queuing (i.e. as long as the bus vehicle decelerates, the bus vehicle can follow the queuing vehicles to drive away from the downstream intersection when the tail vehicle starts, avoiding stopping and queuing at the downstream intersection, as shown in Figure 6 The calculation formula of the guidance speed of the bus vehicle is as follows:

[0129]

[0130] Among them: is the time when the bus vehicle reaches the start point of the speed guidance area, unit: s; γ, η are weight coefficients; L is the distance from the start point of the speed guidance area to the stop line of the downstream intersection, unit: m; V b is the average speed of the bus vehicle, unit: km / h, V min is the minimum speed of the bus vehicle; a d is the deceleration of the bus vehicle, unit: m / s 2 , is the minimum deceleration of the bus vehicle, is the maximum deceleration of the bus vehicle; X (i,j) represents the queuing length of the downstream intersection vehicle at the i phase of the j signal period, unit: m, q is the arrival rate of the social vehicle at the downstream intersection; L v is the average length of the social vehicle at the downstream intersection, is the red light start time at the i phase of the j signal period; T m is the start time of the tail social vehicle, S is the saturation flow rate of the social vehicle at the downstream intersection, is the green light start time at the i phase of the j+1 period.

[0131] When only a bus deceleration guidance strategy is implemented, the buses are traveling at reduced speed, resulting in greater travel delays compared to buses traveling at normal speeds. This travel delay is divided into two parts: first, the delay from the speed guidance area to the end of the queue; and second, the delay following the queue from the end to the intersection. Therefore, the formula for calculating the travel delay time of buses due to deceleration is as follows:

[0132]

[0133] 2) Simultaneously implementing both the bus deceleration guidance strategy and the early red light termination strategy is necessary to ensure that the initiation signal for the next green light cycle reaches the buses, preventing them from stopping and waiting. (That is, simply reducing the bus speed to the minimum speed when the green light turns on in the next cycle is insufficient to ensure the initiation signal reaches the buses, requiring them to still stop and wait. The early red light termination strategy is then required to ensure the initiation signal reaches the buses and prevent them from stopping and waiting.) Figure 7 As shown in the figure, the formula for calculating the guiding speed of the bus at this time is as follows:

[0134] V i-four =V min .

[0135] The formula for calculating the early red light termination time is as follows:

[0136]

[0137] in: T is the red light end time of the i-th phase in the j-th signal period; a This refers to the time it takes for a bus to reach the end of the queue at the designated speed. and The maximum red light cut-off time; G (i-1,j) The original green light duration for the previous phase in the direction of traffic flow; T (i-1,j) The time required for the queue of vehicles in the previous phase of the public transport direction to completely dissipate; r 1j Red light duration for priority phase; q 1j Priority phase social vehicle arrival rate; S 1j The priority phase is the saturation flow rate of social vehicles.

[0138] It is worth noting that, without changing the signal cycle length, early red light termination will shorten the green light time in other phases. To ensure the minimum green light time in the previous phase while avoiding excessive saturation of the previous phase's signal, a maximum early red light termination time needs to be specified.

[0139] When simultaneously implementing a bus deceleration guidance strategy and a red light early termination strategy, the reduced delay for buses after the red light early termination is equal to the sum of the waiting time for the red light before signal timing adjustment and the time it takes for the starting wave to travel to the end of the queue when the green light turns on. Therefore, the formula for calculating the signal delay time of buses in the priority phase caused by the red light early termination is as follows:

[0140]

[0141] Because of the early termination of red lights, vehicles traveling in the same direction as public transportation can also leave the intersection earlier, such as Figure 8 As shown, the formula for calculating the signal delay time of social vehicles in the priority phase caused by the early termination of red lights is as follows:

[0142]

[0143] Furthermore, after implementing early red light termination, the green light time for non-priority phases is compressed. Vehicles that could have left the intersection in the current signal cycle before optimization may now have to wait for the next green light cycle to pass, leading to an increase in the queuing time for non-priority phase vehicles at the intersection. Figure 9 As shown, the formula for calculating the signal delay time of social vehicles in non-priority phases caused by early red light termination is as follows:

[0144]

[0145] Where: q ij For non-priority phase social vehicle arrival rate; S ij This indicates the saturation flow rate of social vehicles in the non-priority phase.

[0146] As shown above, the calculation model for the sum of the travel delay time and signal delay time for each vehicle at the downstream intersection when implementing the bus deceleration guidance strategy and / or the red light early termination strategy is as follows:

[0147]

[0148] Wherein, ΔD' represents the reduction in delays caused by the optimized bus service. O'cc b This represents the average occupancy rate of public transport vehicles; ΔD′ c This indicates the delay caused by the reduction in social vehicles after optimization. Occ c This represents the average occupancy rate of social vehicles; J is the total number of signal phases at the intersection. To optimize the number of buses passing through within a time period; To optimize the total number of social vehicles passing through within a time period.

[0149] Since the problem of minimizing the sum of travel delay time and signal delay time of each vehicle at downstream intersection is a nonlinear problem, a genetic algorithm (GA) is used to solve it. As a self-adjusting global search optimal solution algorithm, it repeatedly simulates the selection, crossover, mutation and other phenomena in natural selection and genetics, and finally decodes the optimal individual in the last generation to obtain the optimal solution that meets the requirements. Based on SUMO simulation software and Python programming software, the combination is realized through Traci interface. Before simulation, the signal timing parameters, traffic flow and other parameters are given, and the signal timing and bus vehicle speed guidance in the simulation are modified in real time through genetic algorithm. The optimal solution that meets the objective function is selected, and finally the vehicle travel delay, queuing delay, signal delay and parking times are calculated. The specific steps are shown in Figure 10 .

[0150] Application example:

[0151] The intersection of Zhengzhou Jinqian Fifth Street and Jinnan Third Road is selected as an example. The main road Jinqian Fifth Street is a six-lane road, and the secondary road Jinnan Third Road is a four-lane road. A simulation diagram is established as shown in Figure 11 .

[0152] The east approach straight lane of the intersection is taken as the research object. The length of the speed guidance area is from the stop line of the intersection to the upstream 200m of the stop line. The bus departure interval is 300s, and only the bus on the trunk road applies for priority. When the bus enters the speed guidance area, the current social vehicle queue length is obtained, and the bus speed guidance algorithm and intersection bus priority application algorithm are activated. The speed of the bus on the road segment is set to 20-40km / h. The signal timing scheme of the intersection is shown in Figure 5 . The intersection flow at different times is collected as shown in Table 1.

[0153] Table 1 Intersection flow at different times

[0154]

[0155] The Traci interface of SUMO simulation software is used to control the simulation experiment to obtain the vehicle speed, distance from the intersection, queue length at the intersection and signal phase, and dynamically adjust the bus speed, signal phase sequence and phase duration to provide priority for the bus approaching the intersection. The simulation period is 4500s, and the simulation accuracy is 1 step / s. Considering the instability of the initial stage of simulation, the data generated may have errors. The first 900s is the simulation warm-up period, and the data during 900-4500s is selected for subsequent analysis. The traffic volume at different times is selected for comparative analysis from the average parking times and average delay time of the bus, and the results are shown in Figure 13 and Figure 14 .

[0156] By Figure 13 It can be seen that under different road traffic, the average number of stops of the bus vehicle is greatly improved after optimization compared with before optimization. Especially under low flow conditions, the bus vehicle after optimization can basically pass through the intersection without stopping; under medium flow conditions, the average number of stops is reduced from 0.67 to 0.25, with a decrease ratio of 62.69%; with the continuous increase of traffic flow, the average number of stops increases significantly, but it is still reduced from 0.75 before optimization to 0.5 after optimization, with a decrease ratio of 33.33%. This is because under low flow, the queue length of social vehicles is short, and the influence on the speed guidance of the bus vehicle and signal optimization is small, therefore, the optimization effect on the average number of stops is particularly significant under low flow environment. Under high flow, the queue length of social vehicles increases correspondingly, resulting in a significant increase in the average number of stops.

[0157] Figure 14 It can be seen that under different flow, the average delay time of the bus vehicle is reduced after optimization compared with before optimization. Under low, medium and high flow, the average delay time is reduced by 69.76%, 43.98% and 22.19% respectively, with the continuous increase of traffic flow, the average delay gradually increases, and the optimization ratio decreases, but the average decrease after optimization is still 45.31% compared with before optimization. Under low flow environment, the number of social vehicles is small, and the green time of each phase is relatively sufficient, which is more conducive to the algorithm to adjust the signal timing and provide signal priority for the bus vehicle. Under medium and high flow environment, the proportion of social vehicles increases, and the space available for speed guidance of the bus vehicle is relatively compressed, the green time available for adjustment of each phase decreases, resulting in a decrease in optimization ratio.

[0158] The above-described embodiments are only preferred embodiments of the present application, merely used to explain the present application, and are not intended to limit the scope of the present application. For those skilled in the art, of course, other embodiments can be easily made by substitution or change based on the technical content disclosed in the present specification, therefore, any changes and improvements made on the principles of the present application shall be included in the scope of the present application.

Claims

1. A time-prediction-based bus lane control method for exclusive bus lane, characterized by, The method comprises the following steps: S1, in a vehicle-network joint environment, obtaining a driving speed of a bus, a real-time distance from the bus to a stop line of a downstream intersection, a signal state of the downstream intersection when the bus reaches a start point of a speed guidance area, and a vehicle passing situation of the downstream intersection, and predicting a time for the bus to reach the start point of the speed guidance area by using an extended Kalman filter model; S2, if the signal state of the downstream intersection when the bus reaches the start point of the speed guidance area is a green light, judging whether the bus can pass through the downstream intersection without stopping by uniformly driving at the current speed according to a remaining time of the green light of the downstream intersection, if yes, maintaining the original driving speed and the original signal timing, if no, executing a bus acceleration guidance strategy and / or a green light extension strategy, and solving a sum of a travel delay time and a signal delay time of each vehicle at the downstream intersection to be minimum by using a genetic algorithm; The execution of the bus acceleration guidance strategy and / or the green light extension strategy specifically includes two cases: 1) only the bus acceleration guidance strategy can make the bus pass through the downstream intersection without stopping, in this case, a calculation formula of a guidance speed of the bus is as follows: In the case of only executing the bus acceleration guidance strategy, a calculation formula of the travel delay time of the bus caused by the acceleration of the bus is as follows: ， wherein: is the time for the bus to reach the start of the speed guidance zone, in seconds; is the end time of the green light for the ith phase in the jth signal cycle, in seconds; is the distance from the start of the speed guidance zone to the stop line of the downstream intersection, in meters; , is the weighting coefficient; is the average speed of the bus, in km / h, and , is the maximum speed allowed for the bus, is the minimum speed for the bus; is the acceleration of the bus, in m / s 2 , and , is the minimum acceleration of the bus, is the maximum acceleration of the bus; 2) the bus acceleration guidance strategy and the green light extension strategy are executed simultaneously to make the bus pass through the downstream intersection without stopping, in this case, a calculation formula of the guidance speed of the bus is as follows: ； In this case, a calculation formula of the green light extension time is as follows: ； In the case of simultaneously executing the bus acceleration guidance strategy and the green light extension strategy, a calculation formula of the signal delay time of the bus in the priority phase caused by the green light extension is as follows: ， wherein: is the remaining green time of the public traffic phase in the jth signal cycle, in seconds; and , , , is the maximum green extension time, in seconds; is the length of the signal cycle of the downstream intersection, in seconds; is the minimum green split of the ith phase; is the minimum green time of the ith phase, in seconds; is the green loss time, in seconds; q is the social vehicle arrival rate of the downstream intersection, and S is the saturated flow rate of the social vehicles of the downstream intersection. A calculation formula of the signal delay time of the social vehicle in the priority phase caused by the green light extension is as follows: ， wherein: red light time for non-preferred phase; A calculation formula of the signal delay time of the social vehicle in the non-priority phase caused by the green light extension is as follows: ， wherein: is the red light duration for the priority phase; is the social vehicle arrival rate for the priority phase; is the social vehicle saturated flow rate for the priority phase; S3, if the signal state of the downstream intersection when the bus reaches the start point of the speed guidance area is a red light, judging whether the bus can reach the downstream intersection without stopping and waiting for the green light of the next signal cycle to light up by uniformly driving at the current speed according to a remaining time of the red light of the downstream intersection, if yes, maintaining the original driving speed and the original signal timing, if no, executing a bus deceleration guidance strategy and / or a red light early break strategy, and solving the sum of the travel delay time and the signal delay time of each vehicle at the downstream intersection to be minimum by using the genetic algorithm; ， wherein: is the non-priority phase social vehicle arrival rate; denotes the non-priority phase social vehicle saturated flow rate; The execution of the bus deceleration guidance strategy and / or the red light early break strategy specifically includes two cases: 1) only the bus deceleration guidance strategy can make a start wave of the green light of the next signal cycle reach the bus to avoid stopping and waiting, in this case, a calculation formula of the guidance speed of the bus is as follows: In the case of only executing the bus deceleration guidance strategy, a calculation formula of the travel delay time of the bus caused by the deceleration of the bus is as follows: ， Wherein: is the time for the bus to arrive at the start of the speed guidance area, unit: s; , is the weight coefficient; is the distance from the start of the speed guidance area to the stop line of the downstream intersection, unit: m; is the average speed of the bus, unit: km / h, is the minimum speed of the bus; is the deceleration of the bus, unit: m / s 2 , , is the minimum deceleration of the bus, is the maximum deceleration of the bus; represents the vehicle queue length at the downstream intersection in the i phase of the j signal cycle, unit: m, , is the arrival rate of social vehicles at the downstream intersection; is the average length of social vehicles at the downstream intersection, is the red light start time in the i phase of the j signal cycle; is the start time of the tail social vehicle, , is the saturation flow rate of social vehicles at the downstream intersection, is the green light start time in the i phase of the j+1 cycle; ​ ； 2) only when the bus speed reduction guidance strategy and the red light early break strategy are executed simultaneously, the start wave can reach the bus to avoid stopping and waiting when the next signal cycle is green, and the bus guidance speed calculation formula is as follows: ； At this time, the calculation formula of the red light early break time is as follows: ， wherein: is the end of red time for the i-th phase in the j-th signal cycle; is the time for the bus to travel to the back of the queue at the lead speed, ; and , , is the maximum red early termination time; is the original green time for a phase in the bus travel direction; is the time for the queue of vehicles in the bus travel direction to completely dissipate; is the red time for the priority phase; is the arrival rate of social vehicles for the priority phase; is the saturated flow rate of social vehicles for the priority phase; When the bus speed reduction guidance strategy and the red light early break strategy are executed simultaneously, the calculation formula of the signal delay time of the bus in the priority phase caused by the red light early break is as follows: ； The calculation formula of the signal delay time of the social vehicle in the priority phase caused by the red light early break is as follows: ； The calculation formula of the signal delay time of the social vehicle in the non-priority phase caused by the red light early break is as follows: ， where: is the non-priority phase social vehicle arrival rate; denotes the non-priority phase social vehicle saturated flow rate.

2. The time prediction-based bus lane priority control method according to claim 1, wherein The vehicle passing situation of the downstream intersection includes the vehicle flow of each passing direction of the intersection and the queue length of the social vehicle when the signal state of the downstream intersection is red.

3. The time prediction-based bus lane priority control method of claim 1, wherein, The length of the vehicle speed guidance area is the distance from the stop line of the downstream intersection to 150-200m upstream of the stop line.

Images