Traffic signal control method and control equipment

Abstract

The invention discloses a traffic signal control method and control equipment, and relates to the technical field of traffic signal control, and the method comprises the steps: a traffic data analysis module calculates a traffic flow density equalization coefficient, a signal period adaptation coefficient and a pedestrian passing influence coefficient based on a standardized traffic data set, and carries out the normalization processing; and inputting the normalized traffic flow density equalization coefficient, the signal period adaptation coefficient and the pedestrian passing influence coefficient into a dynamic traffic flow equalization index model, and outputting a dynamic equalization index. According to the traffic signal timing optimization module, corresponding optimization strategies are started according to different ranges of dynamic balance indexes, hierarchical control is achieved, appropriate timing adjustment modes are adopted under different traffic conditions, quick response during sudden serious congestion is guaranteed, timing can be reasonably optimized during normal or slight congestion, and the traffic safety is improved. And the adaptability and the overall operation efficiency of the traffic signal control system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic signal control, and in particular to a traffic signal control method and control equipment. Background Art

[0002] With the acceleration of urbanization and the continued growth of motor vehicle ownership, urban traffic congestion has become a global problem. Traditional traffic signal control systems, which often rely on fixed-cycle timing schemes, struggle to adapt to dynamic changes in traffic flow. This leads to inefficient intersections, long queues, increased pedestrian wait times, and even increased risk of traffic accidents.

[0003] In recent years, the development of intelligent transportation system technologies has driven the evolution of traffic signal control towards data-driven adaptive control. Existing methods use sensors, cameras, and other devices to collect data such as vehicle flow and speed, and combine them with algorithms such as fuzzy control and neural networks to dynamically adjust signal cycles and phase durations. Some solutions have initially considered the impact of pedestrian waiting time on timing. However, existing technologies generally suffer from insufficient multi-dimensional data integration. For example, they focus solely on vehicle flow data while ignoring key factors such as the frequency of pedestrian illegal crossings and the intensity of crossing demand. Furthermore, they lack systematic quantitative analysis of signal cycle adaptability and vehicle density balance, limiting the global optimization capabilities of control strategies.

[0004] Existing technologies for traffic signal control suffer from numerous deficiencies, specifically a neglect of pedestrian needs, a rigid dynamic response mechanism, and a lack of balance. On the one hand, the lack of quantitative analysis of accumulated pedestrian waiting times, the intensity of crossing demands, and illegal crossing behaviors undermines pedestrian rights and interests, making it difficult to effectively reflect pedestrians' actual needs in signal control. On the other hand, signal adjustments primarily rely on preset thresholds or static models, making it difficult to make timely and appropriate adjustments in complex and changing traffic scenarios such as traffic accidents and temporary road closures, leading to a disconnect between signal control and actual traffic conditions. Furthermore, the imperfect coupling mechanism between the signal cycle adaptation coefficient, traffic flow density, and pedestrian impact can easily lead to local optimization and global imbalance, preventing efficient and balanced traffic flow overall.

[0005] Therefore, it is necessary to invent a traffic signal control method and a control device to solve the above problems. Summary of the Invention

[0006] The object of the present invention is to provide a traffic signal control method and a control device to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a traffic signal control method, comprising the following steps:

[0008] S1. The traffic data acquisition module obtains real-time traffic data of the target intersection, wherein the real-time traffic data includes a traffic flow data set, a signal data set, and a pedestrian data set;

[0009] S2, the traffic data processing module performs denoising and normalization processing on the real-time traffic data to obtain a standardized traffic data set;

[0010] S3. The traffic data analysis module calculates the traffic flow density balance coefficient, the signal cycle adaptation coefficient, and the pedestrian influence coefficient based on the standardized traffic data set, performs normalization processing, and inputs the normalized traffic flow density balance coefficient, the signal cycle adaptation coefficient, and the pedestrian influence coefficient into the dynamic traffic flow balance index model to output the dynamic balance index;

[0011] S4. A strategy generation module generates a signal optimization strategy according to the dynamic equilibrium index;

[0012] S5. The traffic signal timing optimization module adjusts the phase duration and cycle timing plan of the traffic lights at the target intersection based on the signal optimization strategy.

[0013] Preferably, the traffic flow dataset includes instantaneous values of vehicle flow, lane saturation flow rate, current vehicle flow density and historical average vehicle flow density for the same period; the signal dataset includes current signal cycle duration, ideal signal cycle reference value and cycle adjustment frequency; the pedestrian dataset includes accumulated pedestrian waiting time, pedestrian crossing demand intensity and pedestrian illegal crossing frequency.

[0014] Preferably, the traffic flow density balance coefficient is specifically:

[0015]

[0016] Among them, k is the compensation factor of the historical traffic density deviation rate, ε1 is the smoothing factor, n is the total number of lanes, Q i is the instantaneous value of the traffic flow in lane i of the target intersection, S i is the saturation flow rate of lane i, ln is the logarithm with base e, where e is the Euler number, and ρ dev is the historical traffic density deviation rate, and the calculation formula is: Among them, ρ cur is the current traffic density, ρ his It is the average traffic density during the same period in history.

[0017] Preferably, the signal period adaptation coefficient is specifically:

[0018]

[0019] Among them, T c is the current signal cycle length, Tref is the ideal signal period reference value, f adj is the period adjustment frequency, ω1 is the compensation factor for the current signal period, ω2 is the compensation factor for the ideal signal period reference value, ω3 is the compensation factor for the period adjustment frequency, e is the Euler number, ε2 is the smoothing factor, and max is the maximum value function.

[0020] Preferably, the pedestrian traffic impact coefficient is specifically:

[0021]

[0022] Among them, W p is the cumulative waiting time of pedestrians, W0 is the waiting time threshold, D p is the pedestrian crossing demand intensity, F v is the frequency of pedestrian illegal crossing, η1 is the compensation factor for the cumulative amount of pedestrian waiting time, η2 is the compensation factor for the intensity of pedestrian crossing demand, η3 is the compensation factor for the frequency of pedestrian illegal crossing, ε3 is the smoothing factor and ε3>0, lg is the logarithmic function with base 10, τ is the amplification coefficient, and max is the maximum value function.

[0023] Preferably, the dynamic traffic flow equilibrium index model is specifically:

[0024]

[0025] Among them, α′ is the normalized traffic flow density balance coefficient, β′ is the normalized signal cycle adaptation coefficient, γ′ is the normalized pedestrian traffic influence coefficient, △ε is the offset, k1, k2 and k3 are weight coefficients, k1+k2+k3=1 and k1, k2 and k3∈[0,1].

[0026] Preferably, the signal optimization strategy includes:

[0027] Strategy 1: Extend the green light phase duration in the direction of high traffic flow, and adjust the ratio to

[0028] Strategy 2: shorten the phase period when the pedestrian waiting time exceeds the threshold by δ×T c ;

[0029] Strategy three: dynamically insert emergency phases to alleviate sudden congestion, and the insertion frequency is proportional to max(1-δ, 0).

[0030] Preferably, the traffic signal timing optimization module is executed as follows:

[0031] If the dynamic equilibrium index δ<δ1, the first-level optimization strategy is activated, and the second and third strategies described in claim 7 are used in combination;

[0032] If the dynamic equilibrium index δ1≤δ<δ2, the secondary optimization strategy is activated, and the strategy 1 described in claim 7 is used. If the accumulated waiting time of pedestrians exceeds the threshold, the strategies 1 and 2 described in claim 7 are used in combination.

[0033] If the dynamic equilibrium index δ≥δ2, maintain the current strategy and only monitor data updates.

[0034] Preferably, the following modules are included:

[0035] Traffic data collection module, used to obtain real-time traffic data of the target intersection;

[0036] A traffic data processing module, configured to perform denoising and normalization processing on the real-time traffic data to obtain a standardized traffic data set;

[0037] a traffic data analysis module, configured to normalize the traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian influence coefficient calculated from the standardized traffic data set, input the normalized traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian influence coefficient into a dynamic traffic flow balance index model, and output a dynamic balance index;

[0038] A strategy generation module, configured to generate a signal optimization strategy based on the dynamic equilibrium index;

[0039] The traffic signal timing optimization module is used to adjust the phase duration and cycle timing scheme of the traffic lights at the target intersection according to the signal optimization strategy.

[0040] Technical effects and advantages of the present invention:

[0041] 1. This invention achieves dynamic adjustment of intersection signal timing by collecting real-time traffic flow, signal, and pedestrian data and combining it with a dynamic traffic flow equilibrium index model. The system can quickly respond to complex scenarios such as sudden changes in traffic flow and changing pedestrian demand, significantly improving the real-time and adaptability of traffic signal control and alleviating sudden congestion.

[0042] 2. The present invention uses a traffic data analysis module to calculate the traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian impact coefficient, and constructs a dynamic traffic flow balance index model after normalization. This model comprehensively considers traffic flow density, signal cycle adaptation, and pedestrian impact, and can dynamically output a dynamic balance index that reflects the traffic conditions at the intersection. This provides a scientific basis for the generation of signal optimization strategies, making signal control more in line with real-time traffic needs.

[0043] 3. The present invention uses a strategy generation module to generate a signal optimization strategy based on the dynamic equilibrium index. This strategy includes specific measures such as extending the green light phase duration in high-traffic directions, shortening the phase period when pedestrian waiting time exceeds a threshold, and dynamically inserting emergency phases. This can effectively alleviate traffic congestion, improve vehicle traffic efficiency, and take into account pedestrian traffic needs, thereby enhancing the flexibility and effectiveness of traffic signal control.

[0044] 4. The present invention uses the traffic signal timing optimization module to initiate corresponding optimization strategies according to different ranges of the dynamic equilibrium index, thereby achieving hierarchical control and adopting appropriate timing adjustment methods under different traffic conditions. This not only ensures a rapid response in the event of sudden severe congestion, but also reasonably optimizes timing in normal or mild congestion, thereby improving the adaptability and overall operating efficiency of the traffic signal control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The figure is a flow chart of the method steps of the present invention.

[0046] Figure 2 This is the strategy hierarchical control diagram of the present invention.

[0047] Figure 3 This is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] The present invention provides Figure 1 The traffic signal control method shown includes the following steps:

[0050] S1. The traffic data acquisition module obtains real-time traffic data of the target intersection, wherein the real-time traffic data includes a traffic flow data set, a signal data set, and a pedestrian data set;

[0051] Furthermore, in the above technical solution, the traffic flow dataset includes the instantaneous value of vehicle flow, the saturation flow rate of the lane, the current vehicle flow density and the historical average vehicle flow density for the same period; the signal dataset includes the current signal cycle duration, the ideal signal cycle reference value and the cycle adjustment frequency; the pedestrian dataset includes the accumulated pedestrian waiting time, the intensity of pedestrian crossing demand and the frequency of pedestrian illegal crossing.

[0052] What needs to be known is that the instantaneous value of traffic flow is the real-time vehicle passing volume of the lane at the current moment. The target intersection video image is collected by a high-definition camera, and computer vision technology, such as vehicle detection and tracking algorithm, is used to identify and count the number of vehicles passing through each lane in real time; the saturation flow rate of the lane refers to the maximum traffic capacity of the lane under ideal conditions, such as no interference and continuous green lights, which is obtained through historical statistical data; the current traffic density is obtained by collecting the ratio of the number of vehicles in each lane of the target intersection to the lane length through a high-definition camera to calculate the traffic density at the current moment; the average traffic density in the same period in history is obtained by extracting the traffic density data of the same period in history, such as last week or last month, and calculating the average value as a reference value; the current signal cycle length is the total time of a complete cycle of the intersection signal light, that is, the total time for a red, yellow and green light to alternate, directly from the traffic The current cycle setting value is read from the configuration file of the signal control system; the ideal signal cycle reference value is based on the cycle duration of the historical optimal traffic efficiency of the intersection, such as the recommended values of the morning rush hour and the evening rush hour, and the theoretical optimal cycle is calculated by the Webster formula; the cycle adjustment frequency represents the number of adjustments to the signal cycle duration in the past hour, reflecting the dynamic response needs of the intersection, and is obtained through the operation log records of the traffic signal control equipment; the accumulated pedestrian waiting time represents the total waiting time of all pedestrians at the target intersection during the red light period, and is obtained through a high-definition camera combined with a target tracking algorithm; the pedestrian crossing demand intensity represents the pedestrian gathering rate, and is obtained by using a high-definition camera to count the number of pedestrians arriving at the intersection per unit time; the frequency of pedestrian illegal crossing is obtained by using a high-definition camera to count the number of times pedestrians run a red light or fail to pass according to the signal per unit time.

[0053] S2, the traffic data processing module performs denoising and normalization processing on the real-time traffic data to obtain a standardized traffic data set;

[0054] It should be noted that the specific execution process of the traffic data processing module is as follows: first, targeted processing is performed on the noise characteristics of different data sets. In the traffic flow data set, the IQR method is used to detect abnormal values of the instantaneous value of traffic flow, and the missing values are filled by interpolating the mean of adjacent lanes or filling in the historical data of the same period. The traffic density is smoothed with a 5-minute sliding average filter, and the median filter uses a window of 3 cycles to process the signal cycle length to remove short-term fluctuations; in the signal data set, a logical check is performed on the cycle adjustment frequency, and the number of adjustments within 1 hour is limited to ≤20 times. The ideal cycle reference value is removed by the extreme value of the sliding window to eliminate the influence of abnormal time periods; in the pedestrian data set, abnormal records with a single pedestrian waiting time of more than 300 seconds are eliminated, and Poisson distribution fitting is used to filter out false alarms of pedestrian illegal crossing frequency. After denoising, the normalization processing link is entered. For extreme value sensitive parameters such as the instantaneous value of traffic flow, the current signal cycle length, and the cumulative amount of pedestrian waiting time, the minimum-maximum normalization is used. The formula is: The real-time maximum value in the current detection period is used to eliminate the dimension difference; the period adjustment frequency and the violation crossing frequency are logarithmically normalized, and the formula is: Amplify the impact of low-frequency events; Normal distribution parameters, such as the ideal signal period reference value, are standardized using Z-score. Achieve mean and variance normalization. During the processing, it is necessary to ensure that the normalized data meets the domain requirements of subsequent model operations, such as performing interval truncation on the intensity of pedestrian crossing demand to ensure D p -F v +ε3>0. The resulting standardized traffic dataset prevents noise interference and dimensional differences, providing highly consistent and comparable input data for the subsequent calculation of traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian impact coefficient, ensuring the accuracy and robustness of the dynamic traffic flow balance index model.

[0055] S3. The traffic data analysis module calculates the traffic flow density balance coefficient, the signal cycle adaptation coefficient, and the pedestrian influence coefficient based on the standardized traffic data set, performs normalization processing, and inputs the normalized traffic flow density balance coefficient, the signal cycle adaptation coefficient, and the pedestrian influence coefficient into the dynamic traffic flow balance index model to output the dynamic balance index;

[0056] Furthermore, in the above technical solution, the traffic flow density balance coefficient is specifically:

[0057]

[0058] Among them, k is the compensation factor of the historical traffic density deviation rate, ε1 is the smoothing factor, n is the total number of lanes, Q i is the instantaneous value of the traffic flow in lane i of the target intersection, S i is the saturation flow rate of lane i, ln is the logarithm with base e, where e is the Euler number, and ρ dev is the historical traffic density deviation rate, and the calculation formula is: Among them, ρ cur is the current traffic density, ρ his It is the average traffic density during the same period in history.

[0059] It's important to note that the compensation factor k for the historical traffic density deviation rate is determined by analyzing the mapping relationship between historical congestion event data and lane congestion levels. In actual applications, extensive data statistics have shown that in mild congestion, a k value of 0.2 effectively balances the impact of historical traffic density deviation on the traffic flow density equilibrium coefficient; in severe congestion, a k value of 0.8 allows the model to more accurately reflect actual conditions. The smoothing factor ε1 is a constant, typically set to 0.1. This factor is adjusted through experiments on traffic data from the target intersection or similar intersections, using data fitting methods, to ensure that the dynamic equilibrium index output by the model better reflects actual traffic conditions.

[0060] Furthermore, in the above technical solution, the signal period adaptation coefficient is specifically:

[0061]

[0062] Among them, T c is the current signal cycle length, T ref is the ideal signal period reference value, f adj is the period adjustment frequency, ω1 is the compensation factor for the current signal period, ω2 is the compensation factor for the ideal signal period reference value, ω3 is the compensation factor for the period adjustment frequency, e is the Euler number, ε2 is the smoothing factor, and max is the maximum value function.

[0063] It's important to note that the compensation factor ω1 for the current signal cycle duration, the compensation factor ω2 for the ideal signal cycle reference value, and the compensation factor ω3 for the cycle adjustment frequency are derived using data fitting techniques based on historical intersection operation data. The values of these three compensation factors are determined by fitting and analyzing signal cycle data at different time periods and traffic flows to balance the effects of cycle deviation and adjustment frequency on the signal cycle adaptation coefficient. The smoothing factor ε2 is a constant, typically set to 0.1. Specifically, through experiments with traffic data from the target intersection or similar intersections, the factors are adjusted using data fitting methods to ensure that the dynamic equilibrium index output by the model better matches actual traffic conditions.

[0064] Furthermore, in the above technical solution, the pedestrian traffic impact coefficient is specifically:

[0065]

[0066] Among them, W p is the cumulative waiting time of pedestrians, W0 is the waiting time threshold, D p is the pedestrian crossing demand intensity, F vis the frequency of pedestrian illegal crossing, η1 is the compensation factor for the cumulative amount of pedestrian waiting time, η2 is the compensation factor for the intensity of pedestrian crossing demand, η3 is the compensation factor for the frequency of pedestrian illegal crossing, ε3 is the smoothing factor and ε3>0, lg is the logarithmic function with base 10, τ is the amplification coefficient, and max is the maximum value function.

[0067] It should be noted that the compensation factor η1 for the cumulative amount of pedestrian waiting time, the compensation factor η2 for the intensity of pedestrian crossing demand, and the compensation factor η3 for the frequency of pedestrian illegal crossings are determined based on the association between the intersection safety level and monitoring data; the waiting time threshold W0 is set according to the pedestrian flow at the intersection. If the pedestrian flow at the intersection is large and pedestrians arrive continuously, the waiting time is likely to accumulate. In order to ensure the pedestrian traffic experience, the W0 value should be relatively low. For intersections near commercial centers, where the pedestrian flow is extremely large, it is more appropriate to set W0 to 150 to 180 seconds. Pedestrian traffic signals can be given in a timely manner to avoid long waiting times that cause crowds to gather and cross the road illegally. On the contrary, at intersections with low pedestrian flow, the pedestrian arrival interval is long. Setting W0 to 200 to 250 seconds can balance the travel time of pedestrians and vehicles, so as not to affect the vehicle traffic efficiency due to frequent adjustments to the signal lights for a small number of pedestrians; it is set according to the nature of the surrounding land use. At intersections near schools and hospitals, there are many people with mobility difficulties such as the elderly and children, and it takes a long time for people to cross the street. The W0 value is The time interval needs to be appropriately increased, for example, to 220 to 250 seconds, to ensure that this group of people has sufficient time to cross the road safely. Intersections near industrial parks are primarily trafficked by commuters and pedestrians, and pedestrians move relatively quickly. A W0 value of 180 to 200 seconds can meet this requirement. Based on road grade and speed settings, arterial roads have high speeds and heavy traffic volumes. To ensure vehicle continuity and overall road efficiency, a higher W0 value may be appropriate. For arterial roads with speeds of 60 to 80 km / h, W0 may be set to 200 to 250 seconds to reduce the disruption to vehicle traffic caused by frequent pedestrian clearance. For secondary or branch roads with slower speeds and relatively low traffic volumes, a lower W0 value of 150 to 200 seconds can better balance the rights and interests of pedestrians and vehicles. The smoothing factor ε3 is a constant, typically set to 0.1. This factor is adjusted through experiments with traffic data from the target intersection or similar intersections, using data fitting methods, to make the dynamic equilibrium index output by the model more consistent with actual traffic conditions.

[0068] Furthermore, in the above technical solution, the dynamic traffic flow equilibrium index model is specifically:

[0069]

[0070] Among them, α′ is the normalized traffic flow density balance coefficient, β′ is the normalized signal cycle adaptation coefficient, γ′ is the normalized pedestrian influence coefficient, △ε is the offset and △ε>0, k1, k2 and k3 are weight coefficients, k1+k2+k3=1 and k1, k2 and k3∈[0,1].

[0071] It should be noted that the offset Δε is a constant, usually set to 0.01. Through experimental debugging and historical traffic data fitting, the dynamic equilibrium index output by the model is made more consistent with actual traffic conditions.

[0072] S4. A strategy generation module generates a signal optimization strategy according to the dynamic equilibrium index;

[0073] Furthermore, in the above technical solution, the signal optimization strategy includes:

[0074] Strategy 1: Extend the green light phase duration in the direction of high traffic flow, and adjust the ratio to

[0075] Strategy 2: Shorten the phase period where the pedestrian waiting time exceeds the threshold by δ×T c ;

[0076] Strategy 3: Dynamically insert emergency phases to alleviate sudden congestion, with the insertion frequency proportional to max(1-δ, 0).

[0077] It's important to note that the high-traffic direction is the direction of traffic corresponding to one or more lanes at the target intersection with the largest sum of instantaneous traffic flow values, as determined by comparing the instantaneous traffic flow values of each lane. For example, if an intersection has two traffic directions, east-west and north-south, and at a given moment, the sum of the instantaneous traffic flow values for the east-west lanes is 100 vehicles per minute, while for the north-south lanes it is 60 vehicles per minute, then the east-west direction is the current high-traffic direction.

[0078] S5. The traffic signal timing optimization module adjusts the phase duration and cycle timing plan of the traffic lights at the target intersection based on the signal optimization strategy.

[0079] Furthermore, in the above technical solution, the execution mode of the traffic signal timing optimization module is as follows: Figure 2 Shown are:

[0080] If the dynamic equilibrium index δ<δ1, the first-level optimization strategy is activated, and the second and third strategies described in claim 7 are used in combination;

[0081] If the dynamic equilibrium index δ1≤δ<δ2, the secondary optimization strategy is activated, and the strategy 1 described in claim 7 is used. If the accumulated waiting time of pedestrians exceeds the threshold, the strategies 1 and 2 described in claim 7 are used in combination.

[0082] If the dynamic equilibrium index δ≥δ2, maintain the current strategy and only monitor data updates.

[0083] It should be noted that δ1=0.4, δ2=0.7;

[0084] The first-level optimization strategy reflects the poor traffic conditions at this time, which may include severe congestion or sudden traffic anomalies. The second strategy shortens the phase period where the pedestrian waiting time exceeds the threshold, which can reduce pedestrian detention and avoid subsequent problems caused by pedestrian waiting. The third strategy dynamically inserts the emergency phase to directly alleviate sudden congestion. The combination of the two can quickly respond to urgent and complex traffic conditions, simultaneously optimize both pedestrian and vehicle flows, and improve overall traffic efficiency.

[0085] The secondary optimization strategy described above indicates that traffic pressure is high but not yet urgent. Strategy 1, which extends the green light phase duration in high-traffic directions, can directly alleviate pressure in directions with dense traffic. If the accumulated waiting time of pedestrians exceeds a threshold, combining it with Strategy 2 to optimize the phase cycle can take into account both traffic and pedestrian needs, flexibly respond to moderate traffic imbalances, and achieve more refined regulation.

[0086] Furthermore, in the above technical solution, if Figure 3 As shown, it includes the following modules:

[0087] Traffic data collection module, used to obtain real-time traffic data of the target intersection;

[0088] A traffic data processing module, configured to perform denoising and normalization processing on the real-time traffic data to obtain a standardized traffic data set;

[0089] a traffic data analysis module, configured to normalize the traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian influence coefficient calculated from the standardized traffic data set, input the normalized traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian influence coefficient into a dynamic traffic flow balance index model, and output a dynamic balance index;

[0090] A strategy generation module, configured to generate a signal optimization strategy based on the dynamic equilibrium index;

[0091] The traffic signal timing optimization module is used to adjust the phase duration and cycle timing scheme of the traffic lights at the target intersection according to the signal optimization strategy.

[0092] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A traffic signal control method, characterized in that: The following steps are involved: S1. The traffic data acquisition module obtains real-time traffic data of the target intersection, wherein the real-time traffic data includes a traffic flow data set, a signal data set, and a pedestrian data set; S2, the traffic data processing module performs denoising and normalization processing on the real-time traffic data to obtain a standardized traffic data set; S3. The traffic data analysis module calculates the traffic flow density balance coefficient, the signal cycle adaptation coefficient, and the pedestrian influence coefficient based on the standardized traffic data set, performs normalization processing, and inputs the normalized traffic flow density balance coefficient, the signal cycle adaptation coefficient, and the pedestrian influence coefficient into the dynamic traffic flow balance index model to output the dynamic balance index; S4. A strategy generation module generates a signal optimization strategy according to the dynamic equilibrium index; S5. The traffic signal timing optimization module adjusts the phase duration and cycle timing plan of the traffic lights at the target intersection based on the signal optimization strategy.

2. The traffic signal control method according to claim 1, characterized in that: The traffic flow dataset includes instantaneous traffic volume, lane saturation flow rate, current traffic density, and historical average traffic density for the same period; the signal dataset includes the current signal cycle duration, ideal signal cycle reference value, and cycle adjustment frequency; and the pedestrian dataset includes the accumulated pedestrian waiting time, pedestrian crossing demand intensity, and pedestrian illegal crossing frequency.

3. The traffic signal control method according to claim 1, characterized in that: The traffic flow density balance coefficient is specifically: Among them, k is the compensation factor of the historical traffic density deviation rate, ε1 is the smoothing factor, n is the total number of lanes, Q i is the instantaneous value of the traffic flow in lane i of the target intersection, S i is the saturation flow rate of lane i, ln is the logarithm with base e, where e is the Euler number, and ρ dev is the historical traffic density deviation rate, and the calculation formula is: Among them, ρ cur is the current traffic density, ρ his It is the average traffic density during the same period in history.

4. The traffic signal control method according to claim 1, characterized in that: The signal period adaptation coefficient is specifically: Among them, T c is the current signal cycle length, T ref is the ideal signal period reference value, f adj is the period adjustment frequency, ω1 is the compensation factor for the current signal period, ω2 is the compensation factor for the ideal signal period reference value, ω3 is the compensation factor for the period adjustment frequency, e is the Euler number, ε2 is the smoothing factor, and max is the maximum value function.

5. The traffic signal control method according to claim 1, characterized in that: The pedestrian traffic impact coefficient is specifically: Among them, W p is the cumulative waiting time of pedestrians, W0 is the waiting time threshold, D p is the pedestrian crossing demand intensity, F v is the frequency of pedestrian illegal crossing, η1 is the compensation factor for the cumulative amount of pedestrian waiting time, η2 is the compensation factor for the intensity of pedestrian crossing demand, η3 is the compensation factor for the frequency of pedestrian illegal crossing, ε3 is the smoothing factor and ε3>0, lg is the logarithmic function with base 10, τ is the amplification coefficient, and max is the maximum value function.

6. The traffic signal control method according to claim 1, characterized in that: The dynamic traffic flow equilibrium index model is specifically: Among them, α′ is the normalized traffic flow density balance coefficient, β′ is the normalized signal cycle adaptation coefficient, γ′ is the normalized pedestrian traffic influence coefficient, △ε is the offset, k1, k2 and k3 are weight coefficients, k1+k2+k3=1 and k1, k2 and k3∈[0,1].

7. The traffic signal control method according to claim 1, characterized in that: The signal optimization strategy includes: Strategy 1: Extend the green light phase duration in the direction of high traffic flow, and adjust the ratio to Strategy 2: shorten the phase period when the pedestrian waiting time exceeds the threshold by δ×T c ; Strategy three: dynamically insert emergency phases to alleviate sudden congestion, and the insertion frequency is proportional to max(1-δ, 0).

8. The traffic signal control method according to claim 1, characterized in that: The execution mode of the traffic signal timing optimization module is as follows: If the dynamic equilibrium index δ<δ1, the first-level optimization strategy is activated, and the second and third strategies described in claim 7 are used in combination; If the dynamic equilibrium index δ1≤δ<δ2, the secondary optimization strategy is activated, and the strategy 1 described in claim 7 is used. If the accumulated waiting time of pedestrians exceeds the threshold, the strategies 1 and 2 described in claim 7 are used in combination. If the dynamic equilibrium index δ≥δ2, maintain the current strategy and only monitor data updates.

9. The traffic signal control device according to claim 1, characterized in that: Includes the following modules: Traffic data collection module, used to obtain real-time traffic data of the target intersection; A traffic data processing module, configured to perform denoising and normalization processing on the real-time traffic data to obtain a standardized traffic data set; a traffic data analysis module, configured to normalize the traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian influence coefficient calculated from the standardized traffic data set, input the normalized traffic flow density balance coefficient, signal cycle adaptation coefficient, and pedestrian influence coefficient into a dynamic traffic flow balance index model, and output a dynamic balance index; A strategy generation module, configured to generate a signal optimization strategy based on the dynamic equilibrium index; The traffic signal timing optimization module is used to adjust the phase duration and cycle timing scheme of the traffic lights at the target intersection according to the signal optimization strategy.

Images