A data processing system for intelligent traffic monitoring system based on artificial intelligence

Abstract

The present invention relates to the technical field of intelligent traffic monitoring and discloses a data processing system of an intelligent traffic monitoring system based on artificial intelligence, comprising: collecting information of all waiting vehicles at a signal-controlled intersection, the information including historical violation records and waiting positions; predicting the driving risk rate of each waiting vehicle based on the information, identifying dangerous vehicles and safe vehicles in different travel directions according to the driving risk rate, predicting the travel speed of the waiting vehicles at the signal-controlled intersection, and calculating the travel time through the signal-controlled intersection according to the waiting positions; dynamically adjusting the green light duration within a fixed signal cycle based on the distribution of dangerous vehicles going straight and turning left; restricting the right-turning dangerous vehicles to reduce the risk of dangerous vehicles running side by side in adjacent lanes, calculating the risk of dangerous vehicles running side by side in adjacent lanes, and restricting the right-turning dangerous vehicles based on a set parallel risk threshold to reduce traffic conflicts caused by dangerous vehicles.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic monitoring, and in particular to a data processing system of an intelligent traffic monitoring system based on artificial intelligence. Background Art

[0002] At traffic signal intersections, dynamically adjusting green light duration based on vehicle information can effectively improve traffic efficiency. By identifying the number and type of vehicles in different directions and lanes, the system can appropriately allocate travel time in each direction, ensuring that vehicles can pass through the intersection in the shortest possible time, reducing congestion and improving overall traffic flow.

[0003] There are multiple defects in the existing technology in traffic signal control. First, the allocation of green light duration is usually based on fixed cycles or simple vehicle counts, lacking in-depth analysis of vehicle risks, and especially ignoring the special treatment of dangerous vehicles. Secondly, the existing technology is usually unable to distinguish between dangerous vehicles and safe vehicles in different directions of travel, resulting in the mixed traffic of dangerous vehicles and ordinary vehicles, increasing the possibility of traffic accidents. In addition, in the control of right-turning vehicles, the existing technology is often unable to effectively identify the interference of right-turning vehicles on straight-moving vehicles, especially when the right-turn lane and the straight-moving lane are running in parallel, which is prone to traffic conflicts. Finally, the existing technology is unable to notify the driver in real time that he cannot pass through the signal intersection this time, causing some vehicles to hesitate or cut in at the intersection, further increasing the risk of traffic accidents.

[0004] The present invention proposes a data processing system for an intelligent traffic monitoring system based on artificial intelligence. By collecting vehicle historical information, assessing driving risks, dynamically predicting travel time, and intelligently adjusting green light duration, the flexibility and safety of signal control are significantly improved. Summary of the Invention

[0005] The present invention provides a data processing system for an intelligent traffic monitoring system based on artificial intelligence, which is used to promote the solution of the problems mentioned in the above background technology.

[0006] The present invention provides the following technical solutions: a data processing system for an intelligent traffic monitoring system based on artificial intelligence,

[0007] Dynamic vehicle information collection module, used to collect information on all vehicles waiting at signal-controlled intersections, including historical violation records and waiting locations;

[0008] An intelligent risk classification module is used to predict the driving risk rate of each waiting vehicle based on the information, and identify dangerous vehicles and safe vehicles in different travel directions according to the driving risk rate;

[0009] The directions of travel are straight ahead, left turn and right turn respectively;

[0010] The travel time prediction module is used to predict the speed of waiting vehicles at signal-controlled intersections and calculate the travel time through signal-controlled intersections based on the waiting position;

[0011] The green light duration dynamic control module is used to dynamically adjust the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle;

[0012] The right-turn traffic restriction control module is used to restrict dangerous vehicles turning right and reduce the risk of dangerous vehicles driving side by side in adjacent lanes. Specifically, it includes:

[0013] When the straight-moving waiting vehicle passes the signal-controlled intersection, the right-turning waiting vehicle and the straight-moving waiting vehicle travel in adjacent lanes;

[0014] Calculate the risk of dangerous vehicles parallel driving in adjacent lanes, and restrict dangerous vehicles turning right based on the set parallel risk threshold.

[0015] Optionally, the dynamic vehicle information collection module is used to collect information about all vehicles waiting at signal-controlled intersections, including historical violation records and waiting locations, including:

[0016] The signal-controlled intersection includes two main roads intersecting perpendicularly, wherein each main road includes two branch roads with opposite driving directions;

[0017] Each trunk road consists of three roads: a left-turn road, a straight road, and a right-turn road;

[0018] The two main roads are named Main Channel 1 and Main Channel 2 respectively;

[0019] The two branch roads of Main Channel 1 are named Sub-channel 1 and Sub-channel 2, and the two branch roads of Main Channel 2 are named Sub-channel 3 and Sub-channel 4.

[0020] Four groups of traffic lights are set up at the signal-controlled intersection, corresponding to the four branch roads respectively. Each group of traffic lights includes a straight-ahead signal light and a left-turn signal light, which respectively control the straight-ahead and left-turn of waiting vehicles.

[0021] Optionally, the intelligent risk classification module is configured to predict the driving risk rate of each waiting vehicle based on the information, and identify dangerous vehicles and safe vehicles in different travel directions according to the driving risk rate, including:

[0022] Set up roads for each traffic direction. Specifically, the traffic directions of left turn, straight ahead and right turn correspond to the left turn road, straight ahead road and right turn road of the trunk road respectively.

[0023] The historical violation records include the historical violation time and violation type;

[0024] Calculate the driving risk rate of each waiting vehicle, specifically:

[0025] Get the violation time of the i-th violation record of the waiting vehicle ;

[0026] Get the current time ;

[0027] Calculate the time decay weight of the i-th violation , where λ is the set time attenuation coefficient, which is used to control the impact of violation time on the current calculated driving risk rate, and e is an exponent;

[0028] Set a weight for each violation type and obtain the weight of the violation type corresponding to the i-th violation record ;

[0029] Calculate the risk rate of the i-th violation record = , calculate the sum of the risk rates of all violation records in the history of the vehicle, and record the result as the driving risk rate of the vehicle;

[0030] Set a risk rate threshold, and compare the travel risk rate of the waiting vehicle with the risk rate threshold;

[0031] If the driving risk rate is less than the risk rate threshold, the waiting vehicle is a safe vehicle;

[0032] If the driving risk rate is greater than or equal to the risk rate threshold, the waiting vehicle is a dangerous vehicle.

[0033] Optionally, the travel time prediction module is used to predict the travel speed of waiting vehicles at a signal-controlled intersection and calculate the travel time of the signal-controlled intersection based on the waiting position, including:

[0034] For any waiting vehicle:

[0035] Obtain the historical speeds of waiting vehicles passing through the signal-controlled intersection, calculate the sum of all speeds, calculate the average, and record the result as the average speed;

[0036] The speed of the waiting vehicle passing through the signal-controlled intersection is determined to be the average speed;

[0037] Obtain the entrance position of the waiting vehicle through the signal-controlled intersection and calculate the travel distance from the waiting position to the entrance position of the waiting vehicle;

[0038] Calculate the travel distance divided by the travel speed, and record it as the travel time;

[0039] Set a safety factor to increase the passage time of dangerous vehicles;

[0040] If the waiting vehicle is a dangerous vehicle, the passage time is calculated (1+safety factor), the result is used as the final travel time.

[0041] Optionally, the green light duration dynamic control module is used to dynamically control the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle, including:

[0042] The fixed signal cycle is divided into four stages in chronological order;

[0043] The first stage is that subchannel 1 and subchannel 2 go straight at the same time;

[0044] The second stage is when subchannel 1 and subchannel 2 turn left simultaneously;

[0045] The third stage is when subchannel three and subchannel four go straight at the same time;

[0046] The fourth stage is when subchannel three and subchannel four turn left at the same time;

[0047] Get the total green light duration H in the fixed signal cycle, calculate the total green light duration and divide it by 4, and record the result as the initial green light duration of each stage ;

[0048] Get the total number of dangerous vehicles in stage j ;

[0049] Get the total number R of dangerous vehicles at signal-controlled intersections;

[0050] Calculate the green light allocation coefficient for stage j , ;

[0051] Setting the adjustment coefficient , used to control the adjustment range of the green light duration;

[0052] Calculate the green light duration of stage j , .

[0053] Optionally, the green light duration dynamic control module is used to dynamically control the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle, and further includes:

[0054] Update the green light duration of the jth phase to ensure that the total green light duration in the fixed signal cycle remains unchanged;

[0055] After the update, the green light duration of stage j is , ;

[0056] For any sub-channel in stage j, obtain the travel time of each waiting vehicle on the sub-channel, and combine the travel time and green light time Make comparisons;

[0057] Get the duration greater than the green light The waiting vehicles corresponding to the minimum travel time are recorded as marked vehicles;

[0058] The vehicle marked as such is not allowed to pass through the signal-controlled intersection at this time.

[0059] Optionally, the step of calculating the risk of dangerous vehicles running side by side in adjacent lanes and restricting dangerous vehicles turning right based on a set side by side risk threshold includes:

[0060] Establish the right-turn road relationship, specifically:

[0061] The right-turn lane of sub-channel 2 leads to the right-turn lane of sub-channel 4, and the right-turn lane of sub-channel 3 leads to the right-turn lane of sub-channel 2;

[0062] The right-turn lane of sub-channel one leads to the right-turn lane of sub-channel three, and the right-turn lane of sub-channel four leads to the right-turn lane of sub-channel one;

[0063] For the kth stage of straight driving, get the green light duration of the kth stage ;

[0064] Obtain the average speed of all waiting vehicles passing through the signal-controlled intersection, calculate the mean, and record the result as the standard speed ;

[0065] Get any sub-channel in the kth stage and record it as the target sub-channel;

[0066] Obtain the subchannel leading to the right-turn lane of the target subchannel in the right-turn road relationship, and record it as the associated right subchannel.

[0067] Optionally, the step of calculating the risk of dangerous vehicles paralleling in adjacent lanes and restricting dangerous vehicles from turning right based on a set parallel risk threshold may further include:

[0068] calculate , the result is recorded as the right turn distance;

[0069] Obtain the total number of dangerous vehicles turning right within the right-turn distance of the associated right rotor channel, and record it as the total number of right turns;

[0070] The duration of the target sub-channel traffic is less than the green light duration The total number of dangerous vehicles is recorded as the total number of straight vehicles;

[0071] Get the minimum value between the total number of right turns and the total number of straight vehicles, and record it as the maximum number of parallel vehicles;

[0072] Calculate the maximum number of parallel vehicles / (total number of right turns + total number of straights), and record the result as the risk of parallel driving of dangerous vehicles on the associated right rotor channel and the target sub-channel;

[0073] If the risk is greater than or equal to the parallel risk threshold, the number of dangerous vehicles turning right on the associated right rotor channel is restricted, and the number of dangerous vehicles allowed to pass at a time is the total number of right turns. (1-risk).

[0074] The present invention has the following beneficial effects:

[0075] 1. The data processing system of this AI-based intelligent traffic monitoring system collects information about all types of vehicles on the two main roads and their branch roads, building a precise vehicle information database that specifically distinguishes left, straight, and right turns. By rationally dividing sub-channels one through four and combining the traffic light status of each channel, the specific direction of each waiting vehicle is determined. This facilitates the subsequent vehicle risk classification and travel time prediction modules, ensuring real-time awareness of the vehicle distribution in each lane within any given signal cycle.

[0076] 2. The data processing system of this AI-based intelligent traffic monitoring system, through its intelligent risk classification module, accurately assesses the driving risk rate of waiting vehicles based on their historical violation records. This system comprehensively considers the time and type of violation and, through a time decay coefficient, ensures that the impact of older violations on the current risk assessment gradually decreases, thereby improving the timeliness of risk assessments. Furthermore, by assigning weights to violation types, different types of violations have varying influences in the risk assessment, such as a greater impact for serious violations and a smaller impact for minor violations. This effectively distinguishes between dangerous and safe vehicles, providing data support for subsequent green light duration adjustments and parallel driving control.

[0077] 3. The data processing system of this AI-based intelligent traffic monitoring system uses a travel time prediction module to predict the travel time of waiting vehicles at signalized intersections based on their historical speeds. The travel time calculation incorporates the actual waiting position of waiting vehicles, ensuring that the calculated travel time is consistent with actual road conditions. For dangerous vehicles, a safety factor is added to their travel time, ensuring they have sufficient reaction time when passing through the intersection. This not only improves overall traffic safety and efficiency but also provides reliable data support for subsequent green light duration adjustments.

[0078] 4. The data processing system of this AI-based intelligent traffic monitoring system divides the fixed signal cycle into four phases and calculates the initial green light duration for each phase, ensuring a minimum travel time for each direction of traffic. Based on the number of dangerous vehicles in each phase, the green light duration for each phase is dynamically adjusted, ensuring that phases with a high concentration of dangerous vehicles receive longer green light times. This ensures overall traffic flow stability while reducing the risk of traffic conflicts caused by dangerous vehicles, significantly improving intersection safety and traffic efficiency.

[0079] 5. The data processing system of this AI-based intelligent traffic monitoring system, when detecting dangerous vehicles in both the right-turn lane and the through lane, calculates the risk of juxtaposition and, based on a set risk threshold, determines whether to trigger traffic restrictions. If the risk exceeds the threshold, right-turning vehicles are automatically restricted to ensure that dangerous vehicles do not simultaneously enter the intersection. This effectively reduces the risk of traffic conflicts caused by dangerous vehicles juxtaposing each other, while also improving traffic efficiency through dynamic traffic restrictions. When the right-turn lane is overcrowded, dynamic notifications are provided and vehicles are released in batches, thereby improving traffic safety and stability at the intersection.

[0080] 6. The data processing system of this AI-based intelligent traffic monitoring system alerts vehicles marked as unable to pass through signal-controlled intersections, reducing driver hesitation due to uncertainty about whether they can pass, thereby improving traffic efficiency and safety. This not only ensures the stability of overall traffic flow, but also reduces the risk of traffic conflicts caused by dangerous vehicles, significantly improving intersection safety and traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 This is a functional diagram of the modules of the present invention.

[0082] Figure 2 Schematic diagram of a signal-controlled intersection according to the present invention.

[0083] Figure 3 Schematic diagram of three roads of sub-channel 2 of the present invention. DETAILED DESCRIPTION

[0084] 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.

[0085] Example 1, refer to Figure 1 , a data processing system for an intelligent traffic monitoring system based on artificial intelligence, comprising:

[0086] Dynamic vehicle information collection module, used to collect information on all vehicles waiting at signal-controlled intersections, including historical violation records and waiting locations;

[0087] The technical means to collect information about all vehicles waiting at signal-controlled intersections is to install high-resolution cameras at the intersections and use computer vision technology to detect and identify vehicles.

[0088] Collection content: Obtain the license plate number through license plate recognition, and retrieve the historical violation records of the vehicle corresponding to the license plate in combination with the historical database.

[0089] Real-time identification of the current position and queue order of vehicles at the intersection.

[0090] In this embodiment, refer to Figure 2 At the signal-controlled intersection in the city, there are two main roads intersecting vertically, namely Main Channel 1 and Main Channel 2. Each main road has two branch roads:

[0091] Main channel 1, including:

[0092] Sub-channel 1: From north to south, it includes left-turn roads, straight roads and right-turn roads.

[0093] Subchannel 2: From south to north, in this embodiment, refer to Figure 3 , including left-turn roads, straight roads and right-turn roads.

[0094] Main channel 2, including:

[0095] Sub-channel three: from east to west, including left-turn road, straight road and right-turn road.

[0096] Sub-channel 4: From west to east, it includes left-turn roads, straight roads and right-turn roads.

[0097] The signalized intersection features four sets of traffic lights, one for each branch road, controlling straight ahead and left turns. Right-turning vehicles can proceed freely without waiting for a red light, but they should be aware of the risk of colliding with vehicles going straight.

[0098] The system collects real-time information about all vehicles waiting to go at a signal-controlled intersection. For example, there are four vehicles waiting to go, including:

[0099] Vehicle A: In the left-turn lane of sub-channel 1, it has three traffic violation records (one speeding, two red light violations) and is in the second waiting position.

[0100] Vehicle B: In the through lane of sub-channel 2, with no violation record, waiting position is No. 1.

[0101] Vehicle C: In the right-turn lane of sub-channel three, with one speeding record, waiting position is 3rd.

[0102] Vehicle D: In the through lane of sub-channel 4, it has 2 violation records (1 lane change violation, 1 speeding), and its waiting position is 2nd.

[0103] An intelligent risk classification module is used to predict the driving risk rate of each waiting vehicle based on the information, and identify dangerous vehicles and safe vehicles in different travel directions according to the driving risk rate;

[0104] The system predicts the driving risk rate of each waiting vehicle based on the vehicle's historical violation records:

[0105] Vehicle A:

[0106] Traffic violation 1: Speeding. Within 1 year, the violation weight is 0.6, the time decay is 0.8, and the risk rate = 0.6×0.8=0.48.

[0107] Traffic violation 2: Running a red light. Within 2 years, the violation weight is 0.9, the time decay is 0.6, and the risk rate = 0.9×0.6=0.54.

[0108] Traffic violation 3: Running a red light. Within 3 years, the violation weight is 0.9, the time decay is 0.5, and the risk rate = 0.9×0.5=0.45.

[0109] The total risk rate = 0.48 + 0.54 + 0.45 = 1.47 (higher than the set threshold of 1.0, marked as a dangerous vehicle).

[0110] Vehicle B: No violation record, total risk rate = 0 (marked as a safe vehicle).

[0111] Vehicle C:

[0112] Traffic violation 1: Speeding. Within 2 years, the violation weight is 0.6, the time decay is 0.6, and the risk rate = 0.6×0.6=0.36.

[0113] Total risk rate = 0.36 (below the threshold of 1.0, marked as a safe vehicle).

[0114] Vehicle D:

[0115] Violation 1: Lane change violation. Within 1 year, the violation weight is 0.7, the time decay is 0.8, and the risk rate = 0.7×0.8=0.56.

[0116] Traffic violation 2: Speeding. Within 2 years, the violation weight is 0.6, the time decay is 0.6, and the risk rate = 0.6×0.6=0.36.

[0117] Total risk rate = 0.56 + 0.36 = 0.92 (below the threshold of 1.0, marked as a safe vehicle).

[0118] The directions of travel are straight ahead, left turn and right turn respectively;

[0119] The travel time prediction module is used to predict the speed of waiting vehicles at signal-controlled intersections and calculate the travel time through signal-controlled intersections based on the waiting position;

[0120] Obtain the speed of waiting vehicles at signalized intersections based on vehicle-mounted GPS data:

[0121] Principle: Waiting vehicles upload real-time location data through the on-board GPS module or vehicle network terminal.

[0122] Implementation: Get the GPS position of the vehicle at consecutive time points, calculate the position change and time interval, and estimate the speed.

[0123] Vehicle A (turn left): The average speed is 20 km / h, the distance from the waiting position to the intersection is 40 meters, the travel time = 40 / (20×1000 / 3600) = 7.2 seconds (because it is a dangerous vehicle, the safety factor is 0.2), the actual travel time = 7.2×1.2 = 8.64 seconds.

[0124] Vehicle B (going straight): average speed is 30 km / h, waiting distance is 50 meters, and travel time = 50 / (30×1000 / 3600) = 6 seconds.

[0125] Vehicle C (turn right): average speed is 25 km / h, waiting distance is 30 meters, and travel time = 30 / (25×1000 / 3600) = 4.32 seconds.

[0126] Vehicle D (going straight): average speed is 30 km / h, waiting distance is 60 meters, and travel time = 60 / (30×1000 / 3600) = 7.2 seconds.

[0127] The green light duration dynamic control module is used to dynamically adjust the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle;

[0128] Assume that the total signal cycle length H = 120 seconds, divided into four stages (the initial green light duration of each stage is 30 seconds):

[0129] Phase 1: Subchannel 1 and Subchannel 2 go straight.

[0130] Phase 2: Subchannel 1 and Subchannel 2 turn left.

[0131] Phase 3: Subchannel 3 and Subchannel 4 go straight.

[0132] Stage 4: Subchannel 3 and Subchannel 4 turn left.

[0133] Assume that the number of dangerous vehicles at each stage is:

[0134] Phase 1: 10 vehicles;

[0135] Phase 2: 5 vehicles;

[0136] Phase 3: 15 vehicles;

[0137] Phase 4: 20 vehicles;

[0138] The total number of dangerous vehicles R = 10 + 5 + 15 + 20 = 50;

[0139] Calculate the distribution coefficient for each stage: Stage 1: 10 / 50=0.2.

[0140] Second stage: 5 / 50=0.1 seconds.

[0141] Stage 3: 15 / 50=0.3 seconds.

[0142] Stage 4: 20 / 50=0.4 seconds.

[0143] Assuming the adjustment coefficient β = 0.5, the green light duration in each stage is:

[0144] Phase 1: 33 seconds.

[0145] Second stage: 31.5 seconds.

[0146] Stage 3: 34.5 seconds.

[0147] Stage 4: 36 seconds.

[0148] Ensure the total green light duration remains unchanged:

[0149] 33+31.5+34.5+36=135s;

[0150] The adjusted green light durations are:

[0151] Stage 1: (33 / 135)×120=29.33 seconds.

[0152] Second stage: (31.5 / 135)×120=28 seconds.

[0153] Stage 3: (34.5 / 135)×120=30.67 seconds.

[0154] Stage 4: (36 / 135)×120=32 seconds.

[0155] Calculate the travel time of all waiting vehicles in each direction, assuming the following:

[0156] Minimum travel time in stage 1: 28s;

[0157] Minimum travel time in stage 2: 27s;

[0158] Minimum travel time in stage 3: 31s;

[0159] Minimum travel time in stage 4: 30s;

[0160] For stage 3: the passing time is 31s > the green light time is 30.67s, and the vehicle with the passing time of 31s is marked.

[0161] The marked vehicle is notified that it cannot pass through the signal-controlled intersection this time. The system sends a "no passage" instruction to the marked vehicle's on-board terminal through the on-board communication module (such as 4G / 5G communication or vehicle network communication protocol). The content of the instruction includes the specific time period and reason for the prohibition of passage.

[0162] This reduces drivers' hesitation due to uncertainty about whether they can pass and the possibility of traffic accidents during signal changes, thereby improving traffic efficiency and safety. It not only ensures the stability of overall traffic flow, but also reduces the risk of traffic conflicts caused by dangerous vehicles, significantly improving intersection safety and traffic efficiency.

[0163] The right-turn traffic restriction control module is used to restrict dangerous vehicles turning right and reduce the risk of dangerous vehicles driving side by side in adjacent lanes. Specifically, it includes:

[0164] When the straight-moving waiting vehicle passes the signal-controlled intersection, the right-turning waiting vehicle and the straight-moving waiting vehicle travel in adjacent lanes;

[0165] Calculate the risk of dangerous vehicles parallel driving in adjacent lanes, and restrict dangerous vehicles turning right based on the set parallel risk threshold.

[0166] Assume that in phase 3 (straight subpasses three and four):

[0167] The standard speed is 10 m / s. The standard speed here is derived from all waiting vehicles, including those turning left, going straight, and turning right. Since vehicles from all directions converge at signal-controlled intersections, it is assumed that the speed of each waiting vehicle will affect other waiting vehicles and be affected by other waiting vehicles.

[0168] The target subchannel is subchannel three (straight);

[0169] The associated right rotor channel is sub-channel 1 (right rotation);

[0170] Calculate the right turn distance: 30.67×10=306.7m;

[0171] Get information about dangerous vehicles turning right and going straight:

[0172] Total number of vehicles making dangerous right turns within 306.7 meters in Sub-channel 1: 8;

[0173] The total number of dangerous vehicles with a sub-channel three-way travel time of less than 30.67s: 12;

[0174] Calculate the risk of parallel driving:

[0175] Minimum number of vehicles with dangerous right turns and straight driving: 8;

[0176] Calculate the risk of parallel driving: 8 / 20=0.4;

[0177] Determine the restriction:

[0178] Assuming the parallel risk threshold is 0.3 and the current risk is 0.4>0.3, the traffic restriction policy is triggered.

[0179] Dangerous vehicles turning right on the restricted right rotor channel (sub-channel 1):

[0180] The number of dangerous vehicles allowed to pass at one time: 8×(1-0.4)=4.8, approximately equal to 4;

[0181] That is, during this green light cycle, only four dangerous vehicles turning right are allowed to pass through sub-channel one.

[0182] In this embodiment, the system sends a notification to the restricted right-turning vehicle through the vehicle-mounted communication module (such as 4G / 5G communication or vehicle networking protocol), informing it that the right turn is temporarily prohibited.

[0183] The risk factors for dangerous vehicles driving side by side are as follows:

[0184] 1. Superposition of driving behavior risks:

[0185] Multiple dangerous vehicles in a row: Each dangerous vehicle has a higher risk of an accident, and these risks are compounded when driving in a row.

[0186] The probability of unsafe driving behaviors increases: speeding, frequent lane changes, and overtaking are more common among dangerous vehicles, and driving in parallel increases the probability of them affecting each other.

[0187] 2. Mutual interference under limited space:

[0188] Limited lane space: Parallel driving results in less space between vehicles, leaving less room for drivers to react and maneuver.

[0189] Lateral interference: When dangerous vehicles in parallel try to change lanes, turn right, or accelerate through an intersection, lateral collisions are likely to occur.

[0190] 3. Emergency risk transmission effect:

[0191] If a dangerous vehicle loses control, it will affect parallel vehicles: if it suddenly brakes, skids or accelerates, it will be difficult for other dangerous vehicles to avoid it quickly.

[0192] Visual obstruction: Parallel driving will block the driver's vision, especially right-turning vehicles are likely to be blocked by adjacent straight-moving vehicles, leading to accidents.

[0193] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0194] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A data processing system for an intelligent traffic monitoring system based on artificial intelligence, characterized in that: include: Dynamic vehicle information collection module, used to collect information on all vehicles waiting at signal-controlled intersections, including historical violation records and waiting locations; An intelligent risk classification module is used to predict the driving risk rate of each waiting vehicle based on the information, and identify dangerous vehicles and safe vehicles in different travel directions according to the driving risk rate; The directions of travel are straight ahead, left turn and right turn respectively; The travel time prediction module is used to predict the speed of waiting vehicles at signal-controlled intersections and calculate the travel time through signal-controlled intersections based on the waiting position; The green light duration dynamic control module is used to dynamically adjust the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle; The right-turn traffic restriction control module is used to restrict dangerous vehicles turning right and reduce the risk of dangerous vehicles driving side by side in adjacent lanes. Specifically, it includes: When the straight-moving waiting vehicle passes the signal-controlled intersection, the right-turning waiting vehicle and the straight-moving waiting vehicle travel in adjacent lanes; Calculate the risk of dangerous vehicles parallel driving in adjacent lanes, and restrict dangerous vehicles turning right based on the set parallel risk threshold.

2. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 1 is characterized in that: The dynamic vehicle information collection module is used to collect information about all vehicles waiting at signal-controlled intersections, including historical violation records and waiting locations, including: The signal-controlled intersection includes two main roads intersecting perpendicularly, wherein each main road includes two branch roads with opposite driving directions; Each trunk road consists of three roads: a left-turn road, a straight road, and a right-turn road; The two main roads are named Main Channel 1 and Main Channel 2 respectively; The two branch roads of Main Channel 1 are named Sub-channel 1 and Sub-channel 2, and the two branch roads of Main Channel 2 are named Sub-channel 3 and Sub-channel 4. Four groups of traffic lights are set up at the signal-controlled intersection, corresponding to the four branch roads respectively. Each group of traffic lights includes a straight-ahead signal light and a left-turn signal light, which respectively control the straight-ahead and left-turn of waiting vehicles.

3. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 2 is characterized in that: The intelligent risk classification module is used to predict the driving risk rate of each waiting vehicle based on the information, and identify dangerous vehicles and safe vehicles in different travel directions according to the driving risk rate, including: Set up roads for each traffic direction. Specifically, the traffic directions of left turn, straight ahead and right turn correspond to the left turn road, straight ahead road and right turn road of the trunk road respectively. The historical violation records include the historical violation time and violation type; Calculate the driving risk rate of each waiting vehicle, specifically: Get the violation time of the i-th violation record of the waiting vehicle ; Get the current time ; Calculate the time decay weight of the i-th violation , where λ is the set time attenuation coefficient, which is used to control the impact of violation time on the current calculated driving risk rate, and e is an exponent; Set a weight for each violation type and obtain the weight of the violation type corresponding to the i-th violation record ; Calculate the risk rate of the i-th violation record = , calculate the sum of the risk rates of all violation records in the history of the vehicle, and record the result as the driving risk rate of the vehicle; Set a risk rate threshold, and compare the travel risk rate of the waiting vehicle with the risk rate threshold; If the driving risk rate is less than the risk rate threshold, the waiting vehicle is a safe vehicle; If the driving risk rate is greater than or equal to the risk rate threshold, the waiting vehicle is a dangerous vehicle.

4. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 3 is characterized in that: The travel time prediction module is used to predict the travel speed of waiting vehicles at a signal-controlled intersection and calculate the travel time of the signal-controlled intersection based on the waiting position, including: For any waiting vehicle: Obtain the historical speeds of waiting vehicles passing through the signal-controlled intersection, calculate the sum of all speeds, calculate the average, and record the result as the average speed; The speed of the waiting vehicle passing through the signal-controlled intersection is determined to be the average speed; Obtain the entrance position of the waiting vehicle through the signal-controlled intersection and calculate the travel distance from the waiting position to the entrance position of the waiting vehicle; Calculate the travel distance divided by the travel speed, and record it as the travel time; Set a safety factor to increase the passage time of dangerous vehicles; If the waiting vehicle is a dangerous vehicle, the passage time is calculated (1+safety factor), the result is used as the final travel time.

5. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 2 is characterized in that: The green light duration dynamic control module is used to dynamically control the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle, including: The fixed signal cycle is divided into four stages in chronological order; The first stage is that subchannel 1 and subchannel 2 go straight at the same time; The second stage is when subchannel 1 and subchannel 2 turn left simultaneously; The third stage is when subchannel three and subchannel four go straight at the same time; The fourth stage is when subchannel three and subchannel four turn left at the same time; Get the total green light duration H in the fixed signal cycle, calculate the total green light duration and divide it by 4, and record the result as the initial green light duration of each stage ; Get the total number of dangerous vehicles on the two sub-channels in stage j ; Get the total number R of dangerous vehicles at signal-controlled intersections; Calculate the green light allocation coefficient for stage j , ; Setting the adjustment coefficient , used to control the adjustment range of the green light duration; Calculate the green light duration of stage j , .

6. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 5 is characterized in that: The green light duration dynamic control module is used to dynamically control the green light duration based on the distribution of dangerous vehicles going straight and turning left within a fixed signal cycle, and also includes: Update the green light duration of the jth phase to ensure that the total green light duration in the fixed signal cycle remains unchanged; After the update, the green light duration of stage j is , ; For any sub-channel in stage j, obtain the travel time of each waiting vehicle on the sub-channel, and combine the travel time and green light time Make comparisons; Get the duration greater than the green light The waiting vehicles corresponding to the minimum travel time are recorded as marked vehicles; The vehicle marked as such is not allowed to pass through the signal-controlled intersection at this time.

7. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 1 is characterized in that: The method of calculating the risk of dangerous vehicles running side by side in adjacent lanes and restricting dangerous vehicles turning right based on a set side by side risk threshold includes: Establish the right-turn road relationship, specifically: The right-turn lane of sub-channel 2 leads to the right-turn lane of sub-channel 4, and the right-turn lane of sub-channel 3 leads to the right-turn lane of sub-channel 2; The right-turn lane of sub-channel one leads to the right-turn lane of sub-channel three, and the right-turn lane of sub-channel four leads to the right-turn lane of sub-channel one; For the kth stage of straight driving, get the green light duration of the kth stage ; Obtain the average speed of all waiting vehicles passing through the signal-controlled intersection, calculate the mean, and record the result as the standard speed ; Get any sub-channel in the kth stage and record it as the target sub-channel; Obtain the subchannel leading to the right-turn lane of the target subchannel in the right-turn road relationship, and record it as the associated right subchannel.

8. The data processing system of the intelligent traffic monitoring system based on artificial intelligence according to claim 7 is characterized in that: The method of calculating the risk of dangerous vehicles running side by side in adjacent lanes and restricting dangerous vehicles turning right based on a set side by side risk threshold may further include: calculate , the result is recorded as the right turn distance; Obtain the total number of dangerous vehicles turning right within the right-turn distance from the signal-controlled intersection on the associated right rotor channel, and record it as the total number of right turns; Obtain the target sub-channel's passage time to be less than or equal to the green light time The total number of dangerous vehicles is recorded as the total number of straight vehicles; Get the minimum value between the total number of right turns and the total number of straight vehicles, and record it as the maximum number of parallel vehicles; Calculate the maximum number of parallel vehicles / (total number of right turns + total number of straights), and record the result as the risk of parallel driving of dangerous vehicles on the associated right rotor channel and the target sub-channel; If the risk is greater than or equal to the parallel risk threshold, the number of dangerous vehicles turning right on the associated right rotor channel is restricted, and the number of dangerous vehicles allowed to pass at a time is the total number of right turns. (1-risk).

Images