Intelligent intersection signal lamp timing adaptive optimization control system

Abstract

The invention discloses an intelligent intersection signal lamp timing adaptive optimization control system, and the system carries out the adaptive optimization control of a signal lamp at an intersection based on real-time data analysis through a plurality of information collection and processing modules. The system comprises an environment information real-time acquisition module, a driver information acquisition and storage module, a pedestrian information acquisition and storage module and a vehicle information acquisition and storage module, all the modules are connected with a signal lamp timing controller module, and the signal lamp timing controller module comprises a signal lamp control mode decision sub-module and a signal lamp timing decision sub-module. According to the system, the optimization of factors in multiple aspects such as traffic flow, emission, pedestrian traffic safety and the like is realized through the collected information such as vehicle flow, pedestrian number, traffic jam condition, climate data, psychological state of drivers and the like. The signal lamp control mode comprises fixed timing, dynamic timing and self-adaptive timing, and green light, yellow light and red light periods of the signal lamps can be adjusted according to real-time information such as traffic flow difference and traffic jam conditions, so that the traffic efficiency of the intersection is improved, emission is reduced, and pedestrian safety is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and more particularly to an intelligent intersection signal timing adaptive optimization control system. Background Art

[0002] Most of the existing intersection signal control systems adopt fixed cycle or signal timing strategies based on fixed traffic flows, and are unable to respond to changes in traffic flow in real time, resulting in traffic congestion or poor traffic flow in certain directions. In addition, the existing intersection signal control systems also fail to fully consider the comprehensive influence of various factors such as complex traffic flows, pedestrian demands, driver behaviors, and environmental changes, resulting in inaccurate adjustment of signal timing, and thus affecting traffic efficiency and emission control. Especially at complex intersections or during peak hours, the existing signal control systems often cannot meet traffic demands, causing excessive vehicle waiting times and increased safety hazards for pedestrians crossing the street. Therefore, there is an urgent need for a control system that can dynamically optimize signal timing according to multiple factors such as traffic flow, pedestrian demands, special vehicles at intersections, and emission pollution. Summary of the Invention

[0003] The object of the present invention is to overcome the problems existing in the prior art and propose an intelligent intersection signal timing adaptive optimization control system; the system includes a signal timing controller module, an environmental information real-time acquisition module, a driver information acquisition and storage module, a pedestrian information acquisition and storage module, and a vehicle information acquisition and storage module; wherein, the signal timing controller module includes a signal control mode decision sub-module and a signal timing decision sub-module; through the collaborative work of multiple modules, the system adaptively optimizes according to data such as real-time acquired traffic flow, environmental information, and driver behavior, makes intelligent decisions and optimizes the signal timing of traffic lights, realizes the coordination of traffic flow and pedestrian passage, reduces energy waste and pollutant emissions, and thus improves the traffic capacity and safety of intersections.

[0004] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0005] An intelligent intersection signal timing adaptive optimization control system according to the technical solution includes a signal timing controller module, an environmental information real-time acquisition module, a driver information acquisition and storage module, a pedestrian information acquisition and storage module, and a vehicle information acquisition and storage module; the signal timing controller module according to the technical solution is respectively connected to the environmental information real-time acquisition module, the driver information acquisition and storage module, the pedestrian information acquisition and storage module, and the vehicle information acquisition and storage module; the signal timing controller module according to the technical solution includes a signal control mode decision sub-module and a signal timing decision sub-module; the signal control mode decision sub-module according to the technical solution is connected to the signal timing decision sub-module;

[0006] In the technical solution, the intersection is a crossroads, including four traffic flow directions: south, north, east, and west; three sets of vehicle signal lights are configured for each traffic flow direction, namely straight-ahead signal lights, left-turn signal lights, and right-turn signal lights; each traffic flow direction is also equipped with a set of independent pedestrian signal lights. The signal lights described in the technical solution all include green lights, yellow lights, and red lights.

[0007] The environmental information real-time acquisition module described in the technical solution is used to collect the number of vehicles waiting to pass in all lanes of the intersection described in the technical solution, air humidity, environmental temperature, wind speed, special types of the road section area where the intersection described in the technical solution is located, the widths of the four crosswalks of the intersection described in the technical solution, weather warning signals of the road section area where the intersection described in the technical solution is located, and road conditions, and send them to the signal light timing controller module; the road conditions described in the technical solution refer to whether there is a traffic accident at the intersection described in the technical solution currently.

[0008] The driver information acquisition and storage module described in the technical solution is used to collect and store the following information of the drivers of all vehicles waiting to pass through the intersection: risk preference index, driver's reaction time, decision-making ability index, cognitive bias index, traffic cognitive ability factor, driving age, emotional state, average braking acceleration of the vehicle driven by the driver in the past month, and average braking distance of the vehicle driven by the driver in the past month described in the technical solution, and send them to the signal light timing controller module.

[0009] The pedestrian information acquisition and storage module described in the technical solution is used to collect and store the following information of all pedestrians waiting to pass through the intersection: the number of pedestrians waiting to pass through all lanes of the intersection described in the technical solution, pedestrian special needs identity information, and age, and send them to the signal light timing controller module.

[0010] The vehicle information acquisition and storage module described in the technical solution is used to collect and store the following information of all vehicles waiting to pass through the intersection: vehicle type, fuel type, number of forward and stop times, centroid vehicle speed, vehicle age, distance between the centroid position of the vehicle and the vehicle stop line, wheelbase, and longitudinal acceleration, and send them to the signal light timing controller module.

[0011] The signal light control mode decision sub-module described in the technical solution is used to calculate the traffic congestion evaluation factor, the traffic flow difference coefficient, and select the adopted signal light control mode, which specifically includes three signal light control modes: the first timing control mode, the second timing control mode, and the third timing control mode; if the traffic congestion evaluation factor is less than or equal to the first traffic congestion threshold, the first timing control mode is adopted; if the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is less than or equal to the traffic flow difference threshold, the second timing control mode is adopted; if the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is greater than the traffic flow difference threshold, the third timing control mode is adopted; if the traffic congestion evaluation factor is greater than or equal to the second traffic congestion threshold, the third timing control mode is adopted; the first traffic congestion threshold described in the technical solution is selected within the range of 100 to 1000; the second traffic congestion threshold described in the technical solution is selected within the range of 4000 to 10000; the first traffic congestion threshold described in the technical solution is less than the second traffic congestion threshold; the traffic flow difference threshold described in the technical solution is selected within the range of 50 to 2000;

[0012] The signal light control mode decision sub-module described in the technical solution sends the calculated traffic congestion evaluation factor and traffic flow difference coefficient to the signal light timing decision sub-module; the signal light timing decision sub-module described in the technical solution calculates the emission evaluation factor, the pedestrian passage evaluation factor, the vehicle state evaluation factor, and the driver psychological risk factor, and decides the signal cycle of the vehicle signal light and the pedestrian signal light at the intersection described in the technical solution; the signal cycle described in the technical solution is equal to the sum of the duration of the green light, the duration of the yellow light, and the duration of the red light; the emission evaluation factor described in the technical solution is used to evaluate the impact of the vehicles waiting to pass at the intersection on environmental emissions, the pedestrian passage evaluation factor is used to measure the pedestrian passage situation at the intersection, the vehicle state evaluation factor is used to evaluate the state of the vehicles waiting to pass, and the driver psychological risk factor is used to quantify the psychological risk of the driver in traffic decision-making;

[0013] In the first timing mode described in the technical solution, the signal light timing decision sub-module adopts the yellow light flashing control mode. During the entire signal cycle, all the yellow lights at the intersection described in the technical solution flash periodically at a fixed flashing frequency, and the flashing frequency is selected by the applicator between 0.5 Hz and 2 Hz;

[0014] In the second timing mode described in the technical solution, for all vehicle signal lights and pedestrian signal lights at the intersection described in the technical solution, a fixed timing method is adopted. The signal cycle of each signal light is a fixed value, and the signal cycle is selected by the applicator between 60 seconds and 180 seconds; the duration of the green light is selected by the applicator between 20 seconds and 90 seconds; the duration of the yellow light is selected by the applicator between 3 seconds and 5 seconds; the duration of the red light is selected by the applicator between 30 seconds and 90 seconds.

[0015] In the third timing mode described in the technical solution, an adaptive timing method is adopted. The signal cycles of all vehicle signal lights and pedestrian signal lights at the intersection described in the technical solution are calculated by a deep neural network model.

[0016] The input variables of the deep neural network model used to determine the signal cycle in the third timing mode are traffic congestion evaluation factor, traffic flow difference coefficient, emission evaluation factor, pedestrian passing evaluation factor, vehicle state evaluation factor, and driver psychological risk factor. The output variables are the durations of the green light, yellow light, and red light of the 12 groups of signal lights and 4 groups of pedestrian signal lights at the intersection described in the technical solution. The deep neural network model described in the technical solution has a total of 5 layers of networks. The first layer is the input layer, which contains 6 neurons: traffic congestion evaluation factor, traffic flow difference coefficient, emission evaluation factor, pedestrian passing evaluation factor, vehicle state evaluation factor, and driver psychological risk factor; the second layer is the hidden layer, which contains 12 neurons; the third layer is the hidden layer, which contains 14 neurons; the fourth layer is the hidden layer, which contains 10 neurons; the fifth layer is the output layer, which contains 48 neurons: the durations of the green light, yellow light, and red light of the 12 groups of vehicle signal lights and the durations of the green light, yellow light, and red light of the 4 groups of pedestrian signal lights at the intersection described in the technical solution.

[0017] The traffic flow difference coefficient described in the technical solution is calculated by the following formula:

[0018]

[0019] In the formula, σ p represents the traffic flow difference coefficient; Q k is the traffic flow index of the kth lane at the intersection described in the technical solution; P k is the pedestrian flow index of the kth lane at the intersection described in the technical solution; w1 and w2 are the lane traffic flow influence coefficient and lane pedestrian flow influence coefficient respectively; is the average value of the number of vehicles waiting to pass on the four lanes at the intersection described in the technical solution; is the average number of pedestrians waiting to pass through the four lanes at the intersection described in the technical solution;

[0020] The traffic flow index Q of the k-th lane at the intersection described in the technical solution k The specific calculation formula is as follows:

[0021]

[0022] In the formula, N k is the number of vehicles waiting to pass on the k-th lane at the intersection described in the technical solution; t1 is the traffic flow monitoring time length, in hours. If the current time is within the time period from 7:00 to 9:00 or from 17:00 to 19:00, the traffic flow monitoring time length is taken as 0.25 hours, otherwise it is taken as 1 hour;

[0023] The pedestrian flow index P of the k-th lane at the intersection described in the technical solution k The specific calculation formula is as follows:

[0024]

[0025] In the formula, H k is the number of pedestrians waiting to pass through the k-th lane at the intersection described in the technical solution; t1 is the traffic flow monitoring time length;

[0026] The average number of vehicles waiting to pass on the four lanes at the intersection described in the technical solution The specific calculation formula is as follows:

[0027]

[0028] The average number of pedestrians waiting to pass through the four lanes at the intersection described in the technical solution The specific calculation formula is as follows:

[0029]

[0030] The traffic congestion evaluation factor described in the technical solution is calculated by the following formula:

[0031]

[0032] In the formula, T yd represents the traffic congestion evaluation factor; w3 and w4 are the traffic flow influence coefficient of the intersection and the pedestrian flow influence coefficient of the intersection respectively.

[0033] The emission evaluation factor described in the technical solution is calculated by the following formula:

[0034]

[0035] In the formula, Rv represents the emission evaluation factor; N is the total number of vehicles waiting to pass at the intersection described in the technical solution, specifically equal to the sum of the number of vehicles waiting to pass on the four lanes of the intersection described in the technical solution; V sj is the centroid speed of the j-th vehicle waiting to pass at the intersection described in the technical solution, with the unit of meters per second; C j is the vehicle type factor of the j-th vehicle waiting to pass at the intersection described in the technical solution, which is a dimensionless value. If the vehicle type is a heavy machinery vehicle, the vehicle type factor takes a value of 10; if the vehicle type is a passenger car, the vehicle type factor takes a value of 5; if the vehicle type is a special-purpose vehicle or a special vehicle, the vehicle type factor takes a value of 11; if the vehicle type is a light truck, the vehicle type factor takes a value of 5; if the vehicle type is a medium truck, the vehicle type factor takes a value of 8; if the vehicle type is a heavy truck, the vehicle type factor takes a value of 12; if the vehicle type is a bus, the vehicle type factor takes a value of 15; if the vehicle type is a light commercial vehicle, the vehicle type factor takes a value of 9; if the vehicle type is a motorcycle, the vehicle type factor takes a value of 1; if the vehicle type is a trailer, the vehicle type factor takes a value of 0.5; if the vehicle type is other, the vehicle type factor takes a value of 1; G j is the fuel type factor of the j-th vehicle waiting to pass at the intersection described in the technical solution, which is a dimensionless value. If the fuel type is diesel, the fuel type factor takes a value of 20; if the fuel type is gasoline, the fuel type factor takes a value of 15; if the fuel type is hybrid, the fuel type factor takes a value of 10; if the fuel type is liquefied petroleum gas, the fuel type factor takes a value of 5; if the fuel type is natural gas, the fuel type factor takes a value of 3; if the fuel type is hydrogen, the fuel type factor takes a value of 2; if the fuel type is electricity, the fuel type factor takes a value of 1; if the fuel type is other, the fuel type factor takes a value of 5; W t is the red-light waiting time, with the unit of seconds; P tj is the number of times the j-th vehicle waiting to pass at the intersection described in the technical solution has a forward stagnation, which is a dimensionless value; H t is the climate type factor; A xj is the longitudinal acceleration of the j-th vehicle waiting to pass at the intersection described in the technical solution, with the unit of meters per second squared; E aj is the vehicle age of the j-th vehicle waiting to pass at the intersection described in the technical solution, with the unit of years; k1, k2, k3, k4, k5, k6, k7, k8, and k9 are the traffic density influence coefficient, vehicle speed influence coefficient, vehicle type influence coefficient, fuel type influence coefficient, red-light waiting time influence coefficient, forward stagnation number influence coefficient, climate type influence coefficient, longitudinal acceleration influence coefficient, and vehicle age influence coefficient, respectively;

[0036] Climate type factor H tIt can be calculated by the following formula:

[0037]

[0038] In the formula, t a is the air humidity, with the unit of percentage; h a is the ambient temperature, with the unit of degree Celsius; f s is the wind speed, with the unit of meters per second.

[0039] The pedestrian passing evaluation factor described in the technical solution is calculated by the following formula:

[0040]

[0041] In the formula, P c represents the pedestrian passing evaluation factor; N p is the total number of all pedestrians at the intersection described in the technical solution; P ai is the pedestrian age influence factor of the i-th pedestrian, with the unit of years. If the age of the pedestrian is between 0 and 12 years old, the pedestrian age influence factor takes the value of 10. If the pedestrian age is between 13 and 18 years old, the pedestrian age influence factor takes the value of 7. If the pedestrian age is between 19 and 50 years old, the pedestrian age influence factor takes the value of 5. If the pedestrian age is greater than 51 years old, the pedestrian age influence factor takes the value of 12; S pi is the special population influence factor of the i-th pedestrian, which is a dimensionless value. If the pedestrian is a pregnant woman, the special population influence factor takes the value of 6. If the pedestrian is a blind person, the special population influence factor takes the value of 15. If the pedestrian is a wheelchair user, the special population influence factor takes the value of 10. If the pedestrian is other than a pregnant woman, a blind person or a wheelchair user, the special population influence factor takes the value of 0.1. Pregnant women, blind people and wheelchair users all belong to the identity information of special pedestrian needs; L k is the sum of the widths of the four crosswalks at the intersection described in the technical solution, with the unit of meters; S t is the special section influence factor, which is a dimensionless value. If the section area where the intersection described in the technical solution is located is a school, the special section influence factor takes the value of 10. If the section area where the intersection described in the technical solution is located is a hospital, the special section influence factor takes the value of 8. If the section area where the intersection described in the technical solution is located is a railway station, the special section influence factor takes the value of 7. If the section area where the intersection described in the technical solution is located is a pedestrian commercial street, the special section influence factor takes the value of 6. If the section area where the intersection described in the technical solution is located is other than a school, a hospital, a railway station or a pedestrian commercial street, the special section influence factor takes the value of 0.1; W bis the impact factor of bad weather at the intersection described in the technical solution, which is a dimensionless value. If the weather warning for a road section is a blue warning signal, the impact factor of bad weather is 3; if the weather warning for a road section is a yellow warning signal, the impact factor of bad weather is 5; if the weather warning for a road section is an orange warning signal, the impact factor of bad weather is 8; if the weather warning for a road section is a red warning signal, the impact factor of bad weather is 10; if the weather warning for a road section is other than a blue warning signal, a yellow warning signal, an orange warning signal or a red warning signal, the impact factor of bad weather is 2; R a is the total number of traffic accidents that occurred at the intersection described in the technical solution in the past year; k 10 and k 11 and k 12 and k 13 and k 14 are the pedestrian group impact coefficient, the crosswalk width impact coefficient, the special road section impact coefficient, the bad weather impact coefficient, and the road section accident impact coefficient respectively.

[0042] The vehicle status evaluation factor described in the technical solution is calculated by the following formula:

[0043]

[0044] In the formula, S e represents the vehicle status evaluation factor; N is the total number of vehicles waiting to pass at the intersection described in the technical solution, which is specifically equal to the sum of the number of vehicles waiting to pass on the four lanes of the intersection described in the technical solution; P ej is the distance between the centroid position of the j-th vehicle waiting to pass at the intersection described in the technical solution and the vehicle stop line; L wj is the wheelbase of the j-th vehicle waiting to pass at the intersection described in the technical solution, in meters; V sj is the centroid vehicle speed of the j-th vehicle waiting to pass at the intersection described in the technical solution, in meters per second; L k is the sum of the widths of the four crosswalks at the intersection described in the technical solution, in meters; A xj is the longitudinal acceleration of the j-th vehicle waiting to pass at the intersection described in the technical solution, in meters per second squared; E v is the total number of special vehicles at the intersection described in the technical solution; k 15 and k 16 and k 17 and k8 and k 19 are the traffic density impact coefficient, the vehicle position impact coefficient, the vehicle speed-wheelbase impact coefficient, the longitudinal acceleration impact coefficient, and the special vehicle impact coefficient at the intersection respectively.

[0045] The driver psychological risk factor described in the technical solution is calculated by the following formula:

[0046]

[0047] In the formula, D t represents the driver's psychological risk factor; R q is the risk preference index, which is a dimensionless value with a value range from 1 to 10. The risk preference index is used to represent the driver's acceptance of risk when facing traffic decisions. A value of 1 indicates an extremely conservative driver, and a value of 10 indicates a very aggressive driver. The risk preference index can be obtained from the driver's behavior data and psychological assessment questionnaire; T r is the driver's reaction time, with the unit of seconds. The driver's reaction time refers to the time from when the driver perceives a change in traffic signals or other environmental changes to when the driver starts to make a braking or steering reaction. The driver's reaction time can be measured through a simulated driving test; D m is the decision-making ability index, which is a dimensionless value used to quantify the driver's ability to make decisions in a complex traffic environment. The value range is from 1 to 10, and 10 represents the best decision-making ability. The decision-making ability index is obtained through a behavior test; C bias is the cognitive bias index, which is a dimensionless value used to quantify the cognitive bias that occurs during the driver's traffic decision-making process. The value range is from 1 to 10, and 10 represents the highest cognitive bias. The cognitive bias index can be measured through a simulated driving test; E f is the experience level index, and its value is equal to the driver's driving age, with the unit of years. The larger the value, the higher the driver's experience level; D h is the driving habit factor; E s is the emotional state factor, which is a dimensionless value. If the driver's current emotional state is anxiety, the emotional state factor takes a value of 7. If the driver's current emotional state is anger, the emotional state factor takes a value of 10. If the driver's current emotional state is sadness, the emotional state factor takes a value of 5. If the driver's current emotional state is fear, the emotional state factor takes a value of 4. If the driver's current emotional state is other emotional types other than anxiety, anger, sadness, and fear, the emotional state factor takes a value of 1; T c is the traffic cognitive ability factor, which is a dimensionless value. The traffic cognitive ability factor is used to measure the driver's perception, understanding, and reaction abilities in a complex traffic environment. The value range is from 1 to 10, and 10 represents the highest traffic cognitive ability. The traffic cognitive ability factor is obtained by testing the driver's reaction abilities to traffic signals, road condition information, and emergencies; E nv is the road condition index, which is a dimensionless value. The specific value depends on whether there is a traffic accident at the intersection described in the technical solution at present. If there is no traffic accident, the road condition index takes a value of 1. If there is a traffic accident, the road condition index takes a value of 1.5; k 20 k21 , k 22 , k 23 , k 24 , k 25 , k 26 , k 27 and k 28 are the risk preference influence coefficient, reaction time influence coefficient, decision-making ability influence coefficient, cognitive bias influence coefficient, experience level influence coefficient, driving habit influence coefficient, emotional state influence coefficient, traffic cognitive ability influence coefficient, and road condition influence coefficient, respectively;

[0048] Driving habit factor D h is calculated by the following formula:

[0049] D h = k 29 |A xa | 2 + k 30 P ca

[0050] In the formula, A xa is the average braking acceleration of the vehicle driven by the driver in the past month described in the technical solution; P ca is the average braking distance of the vehicle driven by the driver in the past month described in the technical solution; k 29 and k 30 are the braking acceleration influence coefficient and braking distance influence coefficient, respectively.

[0051] The beneficial effects of the present invention are as follows: The intelligent intersection signal timing adaptive optimization control system of the present invention can not only optimize the signal timing according to real-time data, but also adjust the timing according to the behaviors of drivers and pedestrians and environmental factors such as weather and road conditions, effectively improving traffic flow, reducing congestion, ensuring pedestrian safety, and at the same time reducing energy consumption and emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:

[0053] Figure 1 is the system composition block diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0054] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] Refer to Figure 1, an intelligent signal timing adaptive optimization control system for intersections according to the present invention includes a signal timing controller module, a real-time environmental information acquisition module, a driver information acquisition and storage module, a pedestrian information acquisition and storage module, and a vehicle information acquisition and storage module; the signal timing controller module is respectively connected to the real-time environmental information acquisition module, the driver information acquisition and storage module, the pedestrian information acquisition and storage module, and the vehicle information acquisition and storage module; the signal timing controller module includes a signal control mode decision sub-module and a signal timing decision sub-module; the signal control mode decision sub-module is connected to the signal timing decision sub-module;

[0056] The real-time environmental information acquisition module is used to collect the number of vehicles waiting to pass on all lanes of the intersection, air humidity, environmental temperature, wind speed, special types of the road section area where the intersection is located, the widths of the four crosswalks of the intersection, weather warning signals of the road section area where the intersection is located, and road conditions, and send them to the signal timing controller module; the road conditions refer to whether there is a traffic accident at the intersection currently;

[0057] The driver information acquisition and storage module is used to collect and store the following information of the drivers of all vehicles waiting to pass through the intersection: risk preference index, driver's reaction time, decision-making ability index, cognitive bias index, traffic cognitive ability factor, driving age, emotional state, average braking acceleration of the vehicle driven by the driver in the past month, and average braking distance of the vehicle driven by the driver in the past month, and send them to the signal timing controller module;

[0058] The pedestrian information acquisition and storage module is used to collect and store the following information of all pedestrians waiting to pass through the intersection: the number of pedestrians waiting to pass through all lanes of the intersection, pedestrian special needs identity information, and age, and send them to the signal timing controller module;

[0059] The vehicle information acquisition and storage module is used to collect and store the following information of all vehicles waiting to pass through the intersection: vehicle type, fuel type, number of forward and stop times, centroid vehicle speed, vehicle age, distance between the centroid position of the vehicle and the vehicle stop line, wheelbase, and longitudinal acceleration, and send them to the signal timing controller module;

[0060] The described signal light control mode decision sub-module is used to calculate the traffic congestion evaluation factor, the traffic flow difference coefficient, and select the adopted signal light control mode, specifically including three signal light control modes: the first timing control mode, the second timing control mode, and the third timing control mode; if the traffic congestion evaluation factor is less than or equal to the first traffic congestion threshold, the first timing control mode is adopted; if the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is less than or equal to the traffic flow difference threshold, the second timing control mode is adopted; if the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is greater than the traffic flow difference threshold, the third timing control mode is adopted; if the traffic congestion evaluation factor is greater than or equal to the second traffic congestion threshold, the third timing control mode is adopted; the first traffic congestion threshold is selected in the range of 100 to 1000; the second traffic congestion threshold is selected in the range of 4000 to 10000; the first traffic congestion threshold is less than the second traffic congestion threshold; the traffic flow difference threshold is selected in the range of 50 to 2000;

[0061] The signal light control mode decision sub-module sends the calculated traffic congestion evaluation factor and traffic flow difference coefficient to the signal light timing decision sub-module; the signal light timing decision sub-module calculates the emission evaluation factor, the pedestrian passing evaluation factor, the vehicle state evaluation factor, and the driver psychological risk factor, and determines the signal cycle of the vehicle signal light and the pedestrian signal light at the intersection; the signal cycle is equal to the sum of the duration of the green light, the duration of the yellow light, and the duration of the red light. The emission evaluation factor is used to evaluate the impact of the vehicles waiting to pass at the intersection on environmental emissions, the pedestrian passing evaluation factor is used to measure the pedestrian passing situation at the intersection, the vehicle state evaluation factor is used to evaluate the state of the vehicles waiting to pass, and the driver psychological risk factor is used to quantify the psychological risk of the driver in traffic decision-making.

[0062] The above description is only a preferred embodiment of the present invention, for the purpose of explanation and illustration, and is not a limitation on the present invention itself. The present invention is not limited to the specific embodiments disclosed herein, but is determined by the following claims. Additionally, the descriptions related to specific embodiments in the foregoing cannot be construed as limitations on the scope of the present invention or the definitions of the terms used in the claims. Various other different embodiments and various different deformations of the disclosed embodiments are obvious to those skilled in the art. But all such embodiments, changes, and deformations that do not depart from the basic concept of the present invention are within the scope of the appended claims.

Claims

1. An intelligent intersection signal light timing adaptive optimization control system, characterized in that: It includes a signal light timing controller module, an environmental information real-time acquisition module, a driver information acquisition and storage module, a pedestrian information acquisition and storage module, and a vehicle information acquisition and storage module; the signal light timing controller module is respectively connected to the environmental information real-time acquisition module, the driver information acquisition and storage module, the pedestrian information acquisition and storage module, and the vehicle information acquisition and storage module; the signal light timing controller module includes a signal light control mode decision submodule and a signal light timing decision submodule; the signal light control mode decision submodule is connected to the signal light timing decision submodule; The intersection is a cross intersection, including four traffic flow directions: south, north, east and west; each traffic flow direction is equipped with three sets of vehicle signal lights, namely, a straight signal light, a left turn signal light and a right turn signal light; each traffic flow direction is also equipped with a set of independent pedestrian signal lights, and the signal lights include green lights, yellow lights and red lights; The real-time environmental information acquisition module is used to collect the number of vehicles waiting to pass in all lanes of the intersection, air humidity, ambient temperature, wind speed, the special type of the road section area where the intersection is located, the width of the four crossing strips of the intersection, the weather warning signal of the road section area where the intersection is located, and the road conditions, and send them to the signal light timing controller module; the road conditions refer to whether there is a traffic accident at the intersection at present; The driver information collection and storage module is used to collect and store the following information of drivers of all vehicles waiting to pass through the intersection: risk preference index, driver's reaction time, decision-making ability index, cognitive bias index, traffic cognitive ability factor, driving age, emotional state, average braking acceleration of the driver's driving vehicle in the past month, and average braking distance of the driver's driving vehicle in the past month, and send it to the signal light timing controller module; The pedestrian information collection and storage module is used to collect and store the following information of all pedestrians waiting to pass through the intersection: the number of pedestrians waiting to pass through all lanes at the intersection, the identity information of pedestrians with special needs, and their ages, and send it to the signal light timing controller module; The vehicle information collection and storage module is used to collect and store the following information of all vehicles waiting to pass through the intersection: vehicle type, fuel type, number of travel stops, center of mass speed, vehicle age, distance between the center of mass position of the vehicle and the vehicle stop line, wheelbase, longitudinal acceleration, and send it to the signal light timing controller module; The signal light control mode decision submodule is used to calculate the traffic congestion evaluation factor, the traffic flow difference coefficient, and select the signal light control mode to be adopted, which specifically includes three signal light control modes: the first timing control mode, the second timing control mode and the third timing control mode; if the traffic congestion evaluation factor is less than or equal to the first traffic congestion threshold, the first timing control mode is adopted; if the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is less than or equal to the traffic flow difference threshold, the second timing control mode is adopted; if the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is less than or equal to the traffic flow difference threshold, the second timing control mode is adopted; If the traffic congestion evaluation factor is between the first traffic congestion threshold and the second traffic congestion threshold, and the traffic flow difference coefficient is greater than the traffic flow difference threshold, the third timing control mode is adopted; if the traffic congestion evaluation factor is greater than or equal to the second traffic congestion threshold, the third timing control mode is adopted; the first traffic congestion threshold is selected in the range of 100 to 1000; the second traffic congestion threshold is selected in the range of 4000 to 10000; the first traffic congestion threshold is less than the second traffic congestion threshold; the traffic flow difference threshold is selected in the range of 50 to 2000; The signal light control mode decision submodule sends the calculated traffic congestion evaluation factor and traffic flow difference coefficient to the signal light timing decision submodule; The signal light timing decision submodule calculates the emission evaluation factor, the pedestrian traffic evaluation factor, the vehicle state evaluation factor, and the driver's psychological risk factor, and decides the signal cycle of the vehicle signal light and the pedestrian signal light at the intersection; the signal cycle is equal to the sum of the duration of the green light, the duration of the yellow light, and the duration of the red light; the emission evaluation factor is used to evaluate the impact of vehicles waiting to pass through the intersection on environmental emissions, the pedestrian traffic evaluation factor is used to measure the pedestrian traffic situation at the intersection, the vehicle state evaluation factor is used to evaluate the state of vehicles waiting to pass, and the driver's psychological risk factor is used to quantify the driver's psychological risk in traffic decision-making; The signal light timing decision submodule adopts a yellow light flashing control mode in the first timing mode. During the entire signal cycle, all yellow lights at the intersection flash periodically at a fixed flashing frequency, and the flashing frequency is selected by the user between 0.5 Hz and 2 Hz. In the second timing mode, the signal light timing decision submodule adopts a fixed timing method for all vehicle signal lights and pedestrian signal lights at the intersection, and the signal cycle of each signal light is a fixed value, and the signal cycle is selected by the user between 60 seconds and 180 seconds; the duration of the green light is selected by the user between 20 seconds and 90 seconds; the duration of the yellow light is selected by the user between 3 seconds and 5 seconds; the duration of the red light is selected by the user between 30 seconds and 90 seconds; The signal light timing decision submodule adopts an adaptive timing method in the third timing mode, and the signal periods of all vehicle signal lights and pedestrian signal lights at the intersection are calculated by a deep neural network model; The input variables of the deep neural network model for determining the signal period under the third timing mode are traffic congestion evaluation factor, traffic flow difference coefficient, emission evaluation factor, pedestrian traffic evaluation factor, vehicle state evaluation factor, and driver psychological risk factor, and the output variables are the duration of the green light, the duration of the yellow light, and the duration of the red light of the 12 groups of signal lights and the 4 groups of pedestrian signal lights at the intersection. The deep neural network model has a total of 5 layers of networks, the first layer is the input layer, which contains 6 neurons: traffic congestion evaluation factor, traffic flow difference coefficient, emission evaluation factor, pedestrian traffic evaluation factor, vehicle state evaluation factor, and driver psychological risk factor; The second layer is a hidden layer, which includes 12 neurons; the third layer is a hidden layer, which includes 14 neurons; the fourth layer is a hidden layer, which includes 10 neurons; the fifth layer is an output layer, which includes 48 neurons: the duration of the green light, the duration of the yellow light, the duration of the red light of the 12 groups of vehicle signal lights at the intersection and the duration of the green light, the duration of the yellow light, the duration of the red light of the 4 groups of pedestrian signal lights; The traffic flow difference coefficient is calculated using the following formula: In the formula, σ p represents the traffic flow difference coefficient; Q k is the traffic flow index of the kth lane of the intersection; P k is the pedestrian flow index of the kth lane of the intersection; w1 and w2 are lane vehicle flow influence coefficient and lane pedestrian flow influence coefficient respectively; is the average number of vehicles waiting to pass on the four lanes of the intersection; is the average number of pedestrians waiting to cross the four lanes at the intersection; The traffic flow index Q of the kth lane of the intersection k The specific calculation formula is as follows: Where N k is the number of vehicles waiting to pass on the kth lane of the intersection; t1 is the traffic monitoring time length, in hours. If the current time is between 7:00 and 9:00, or between 17:00 and 19:00, the traffic monitoring time length is 0.25 hours, otherwise it is 1 hour; The pedestrian flow index P of the kth lane of the intersection k The specific calculation formula is as follows: In the formula, H k is the number of pedestrians waiting to pass through the kth lane at the intersection; t1 is the length of traffic monitoring time; The average number of vehicles waiting to pass on the four lanes of the intersection The specific calculation formula is as follows: The average number of pedestrians waiting to cross the four lanes at the intersection The specific calculation formula is as follows: The traffic congestion evaluation factor is calculated using the following formula: Where, T yd represents the traffic congestion evaluation factor; w3 and w4 are the intersection vehicle flow impact coefficient and intersection pedestrian flow impact coefficient respectively.

2. The intelligent intersection signal light timing adaptive optimization control system as claimed in claim 1, characterized in that: The emission assessment factor is calculated using the following formula: In the formula, R v represents the emission evaluation factor; N is the total number of vehicles waiting to pass through the intersection, which is specifically equal to the sum of the number of vehicles waiting to pass on the four lanes of the intersection; V sj is the centroid speed of the jth vehicle waiting to pass at the intersection, in meters per second; C j is the vehicle type factor of the jth vehicle waiting to pass at the intersection, which is a dimensionless value. If the vehicle type is a heavy machinery vehicle, the vehicle type factor is 10; if the vehicle type is a passenger car, the vehicle type factor is 5; if the vehicle type is a special vehicle or a special vehicle, the vehicle type factor is 11; if the vehicle type is a light truck, the vehicle type factor is 5; if the vehicle type is a medium truck, the vehicle type factor is 8; if the vehicle type is a heavy truck, the vehicle type factor is 12; if the vehicle type is a bus, the vehicle type factor is 15; if the vehicle type is a light commercial vehicle, the vehicle type factor is 9; if the vehicle type is a motorcycle, the vehicle type factor is 1; if the vehicle type is a trailer, the vehicle type factor is 0.5; if the vehicle type is other, the vehicle type factor is 1; G j is the fuel type factor of the jth vehicle waiting to pass at the intersection, which is a dimensionless value. If the fuel type is diesel, the fuel type factor is 20; if the fuel type is gasoline, the fuel type factor is 15; if the fuel type is hybrid, the fuel type factor is 10; if the fuel type is liquefied petroleum gas, the fuel type factor is 5; if the fuel type is natural gas, the fuel type factor is 3; if the fuel type is hydrogen, the fuel type factor is 2; if the fuel type is electricity, the fuel type factor is 1; if the fuel type is other, the fuel type factor is 5; W t is the waiting time for the traffic light, in seconds; P tj is the number of times the jth vehicle waiting to pass stops at the intersection, which is a dimensionless value; H t is the climate type factor; A xj is the longitudinal acceleration of the jth vehicle waiting to pass at the intersection, in meters per second squared; E aj is the age of the jth vehicle waiting to pass at the intersection, in years; k1, k2, k3, k4, k5, k6, k7, k8 and k9 are the vehicle quantity influence coefficient, vehicle speed influence coefficient, vehicle type influence coefficient, fuel type influence coefficient, traffic light waiting time influence coefficient, travel stop number influence coefficient, climate type influence coefficient, longitudinal acceleration influence coefficient and vehicle age influence coefficient respectively; Climate type factor H t The following formula can be used for calculation: Where, t a is the air humidity in percentage; h a is the ambient temperature in degrees Celsius; f s is the wind speed in meters per second.

3. The intelligent intersection signal light timing adaptive optimization control system as claimed in claim 1, characterized in that: The pedestrian traffic evaluation factor is calculated using the following formula: Where P c represents the pedestrian traffic evaluation factor; N p is the total number of pedestrians at the intersection; P ai is the pedestrian age impact factor of the ith pedestrian, in years. If the pedestrian is between 0 and 12 years old, the pedestrian age impact factor is 10; if the pedestrian is between 13 and 18 years old, the pedestrian age impact factor is 7; if the pedestrian is between 19 and 50 years old, the pedestrian age impact factor is 5; if the pedestrian is older than 51 years old, the pedestrian age impact factor is 12; S pi is the special population impact factor of the ith pedestrian, which is a dimensionless value. If the pedestrian is a pregnant woman, the special population impact factor is 6; if the pedestrian is a blind person, the special population impact factor is 15; if the pedestrian is a wheelchair user, the special population impact factor is 10; if the pedestrian is a pregnant woman, a blind person or a person other than a wheelchair user, the special population impact factor is 0.

1. Pregnant women, blind people and wheelchair users all belong to pedestrian special needs identity information; L k is the sum of the widths of the four crossing strips at the intersection, in meters; S t is the special section impact factor, which is a dimensionless value. If the intersection is located in a school, the special section impact factor is 10; if the intersection is located in a hospital, the special section impact factor is 8; if the intersection is located in a railway station, the special section impact factor is 7; if the intersection is located in a pedestrian shopping street, the special section impact factor is 6; if the intersection is located in a school, hospital, railway station or other road section other than a pedestrian shopping street, the special section impact factor is 0.1; W b is the inclement weather impact factor of the intersection, which is a dimensionless value. If the weather warning of the section is a blue warning signal, the inclement weather impact factor is 3; if the weather warning of the section is a yellow warning signal, the inclement weather impact factor is 5; if the weather warning of the section is an orange warning signal, the inclement weather impact factor is 8; if the weather warning of the section is a red warning signal, the inclement weather impact factor is 10; if the weather warning of the section is a weather warning signal other than a blue warning signal, a yellow warning signal, an orange warning signal or a red warning signal, the inclement weather impact factor is 2; R a is the total number of traffic accidents that occurred at the intersection in the past year; k 10 , k 11 , k 12 , k 13 and k 14 They are the pedestrian group impact coefficient, the crossing strip width impact coefficient, the special road section impact coefficient, the bad weather impact coefficient, and the road section accident impact coefficient.

4. The intelligent intersection signal light timing adaptive optimization control system as claimed in claim 1, characterized in that: The vehicle status evaluation factor is calculated using the following formula: In the formula, S e represents the vehicle status evaluation factor; N is the total number of vehicles waiting to pass through the intersection, which is specifically equal to the sum of the number of vehicles waiting to pass on the four lanes of the intersection; P ej L is the distance between the center of mass of the jth vehicle waiting to pass at the intersection and the vehicle stop line; wj is the wheelbase of the jth vehicle waiting to pass at the intersection, in meters; V sj is the centroid speed of the jth vehicle waiting to pass at the intersection, in meters per second; L k is the sum of the widths of the four crossing strips at the intersection, in meters; A xj is the longitudinal acceleration of the jth vehicle waiting to pass at the intersection, in meters per second squared; E v is the total number of special vehicles at the intersection; k 15 , k 16 , k 17 , k8 and k 19 They are the vehicle quantity influence coefficient, vehicle position influence coefficient, vehicle speed wheelbase influence coefficient, longitudinal acceleration influence coefficient, and special vehicle influence coefficient at intersections.

5. The intelligent intersection signal light timing adaptive optimization control system as claimed in claim 1, characterized in that: The driver's psychological risk factor is calculated using the following formula: Where D t represents the driver's psychological risk factor; R q is the risk preference index, which is a dimensionless value ranging from 1 to 10. The risk preference index is used to indicate the driver's acceptance of risk when facing traffic decisions, where a value of 1 indicates an extremely conservative driver and a value of 10 indicates a very aggressive driver. The risk preference index can be obtained from driver behavior data and psychological assessment questionnaires; T r The driver's reaction time is measured in seconds. The driver's reaction time refers to the time from when the driver perceives a change in traffic signals or other environmental changes to when he or she begins to brake or turn. The driver's reaction time can be measured through a simulated driving test; D m is the decision-making ability index, which is a dimensionless value used to quantify the driver's ability to make decisions in a complex traffic environment. The value range is 1 to 10, with 10 indicating the best decision-making ability. The decision-making ability index is obtained through behavioral testing; C bias is the cognitive bias index, which is a dimensionless value used to quantify the cognitive bias of drivers in the process of traffic decision-making. The value range is 1 to 10, with 10 indicating the highest cognitive bias. The cognitive bias index can be measured through simulated driving tests; E f It is the experience level index, the value is equal to the driver's driving age, the unit is year, the larger the value is, the higher the driver's experience level is; D h is the driving habit factor; E s is the emotional state factor, which is a dimensionless value. If the driver's current emotional state is anxiety, the emotional state factor is 7; if the driver's current emotional state is anger, the emotional state factor is 10; if the driver's current emotional state is sadness, the emotional state factor is 5; if the driver's current emotional state is fear, the emotional state factor is 4; if the driver's current emotional state is other than anxiety, anger, sadness, and fear, the emotional state factor is 1; T c is the traffic cognition factor, which is a dimensionless value. The traffic cognition factor is used to measure the driver's perception, understanding and reaction ability in a complex traffic environment. The value range is 1 to 10, where 10 represents the highest traffic cognition ability. The traffic cognition factor is obtained by testing the driver's reaction ability to traffic signals, road conditions, and emergencies; E nv is the road condition index, which is a dimensionless value. The specific value depends on whether there is a traffic accident at the intersection. If there is no traffic accident, the road condition index is 1. If there is a traffic accident, the road condition index is 1.

5. 20 , k 21 , k 22 , k 23 , k 24 , k 25 , k 26 , k 27 and k 28 They are risk preference influence coefficient, reaction time influence coefficient, decision-making ability influence coefficient, cognitive bias influence coefficient, experience level influence coefficient, driving habit influence coefficient, emotional state influence coefficient, traffic cognitive ability influence coefficient, and road condition influence coefficient; Driving habit factor D h The calculation is done using the following formula: D h =k 29 |A xa | 2 +k 30 P ca In the formula, A xa is the average braking acceleration of the vehicle driven by the driver in the past month; ca k is the average braking distance of the vehicle driven by the driver in the past month; 29 and k 30 They are the braking acceleration influence coefficient and the braking distance influence coefficient respectively.

Images