Traffic road congestion prediction method and system based on artificial intelligence
Through the traffic road congestion prediction method based on artificial intelligence, combined with historical traffic data and pedestrian crossing video data, a left turn speed model for left turn vehicles is constructed, which solves the problem of traffic congestion caused by pedestrian traffic in traditional methods that cannot accurately predict the traffic congestion problem of left turn vehicles due to pedestrian traffic, and realizes the Accurate prediction and relief of traffic road congestion.
Patent Information
- Application Number
- CN202411770309.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-04
- Publication Date
- 2025-02-18
AI Technical Summary
Traditional traffic road congestion prediction methods cannot accurately determine the traffic congestion caused by left-turning vehicles due to pedestrians failing to pass through pedestrians in time.
Using the traffic road congestion prediction method based on artificial intelligence, the left turn speed model of the left turn vehicle is constructed by obtaining the historical traffic data and pedestrian crossing video data of the target traffic road, and combining the historical congestion value and the comprehensive pedestrian congestion value, the road congestion level is analyzed. .
Accurate prediction of traffic road congestion has been achieved, especially when left-turning vehicles are affected by pedestrian traffic, they can predict and alleviate traffic pressure in advance.
Smart Images
Figure CN119479306A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of traffic roads, and specifically relates to a traffic road congestion prediction method and system based on artificial intelligence. Background Art
[0002] Traffic roads refer to the infrastructure for various vehicles and pedestrians to pass through, and they are an integral part of urban infrastructure construction. According to their status, role, traffic function, and service function to buildings along the roads in the urban road system, as the vehicle ownership rate increases, road congestion is becoming more and more common. As road congestion leads to a significant increase in commuting time, the time available for work (production) is reduced, causing economic losses to drivers and regions. Traffic congestion not only affects individuals, but may also hinder business activities and reduce logistics efficiency, thus having a negative impact on the economy. Therefore, the prediction of traffic road congestion is particularly important. It is an important part of urban traffic management. It helps to plan traffic resources and formulate response strategies in advance, thereby alleviating traffic pressure and improving road traffic efficiency. At present, when the green light for going straight on a traffic road goes out, if pedestrians fail to cross the crosswalk in time, but the green light for turning left is on at this time, left-turning vehicles must give way to pedestrians before passing through the crosswalk after turning left, which will cause left-turning vehicles to be unable to pass in time and cause congestion. Traditional prediction methods can only play lane congestion, but cannot accurately judge the traffic congestion prediction in this case; To this end, the present invention proposes a traffic congestion prediction method and system based on artificial intelligence. Summary of the invention
[0003] The purpose of the present invention is to propose a traffic congestion prediction method and system based on artificial intelligence to solve the problems raised in the above background technology.
[0004] In the first aspect, in order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions: The traffic congestion prediction method based on artificial intelligence is as follows: Step S1, obtaining historical traffic data corresponding to the target traffic road; Step S2, analyzing the historical traffic data of the target traffic road to obtain the historical congestion value of the target traffic road; Step S3, constructing a left-turn speed model for left-turning vehicles corresponding to the target traffic road according to historical traffic data corresponding to the target traffic road; Step S4, obtaining the pedestrian crossing video data of the target traffic road, and analyzing the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road; Step S5, combining the historical congestion value of the target traffic road with the comprehensive pedestrian congestion value to obtain the road congestion level corresponding to the target traffic road.
[0005] Furthermore, the historical traffic data includes the daily traffic flow of the target road in the same month as the current time in the previous year, and the historical average vehicle speed and historical average vehicle distance of the target traffic road in the same month as the current time in the previous year.
[0006] Furthermore, the step S2 includes the following sub-steps: Step S21, acquiring historical traffic data of the target traffic road, obtaining the historical average vehicle speed LSD, the historical average vehicle spacing LCJ and the road daily traffic flow RLL of the target traffic road in the same month as the current time in the previous year; Step S22, obtaining the maximum traffic volume TXL of the target traffic road; Step S23, calculating the historical congestion value LSY of the target traffic road by the formula LSY=1 / LSD+e / LCJ+RLL / TXL; wherein e is a natural constant.
[0007] Furthermore, the step S3 includes the following sub-steps: Step S31, obtaining the left turn arc value corresponding to the left turn operation performed by the left-turning vehicle on the target traffic road on the same day, the lateral distance between the left-turning vehicle and the road edge, and the turning radius of the left-turning vehicle when turning; Step S32, constructing a left-turn speed model for the left-turn vehicle, defining the point where the left-turn vehicle has the minimum speed during the turning process as the speed demarcation point; the left-turn speed model is divided into two parts, a deceleration part JS(t) and a speed-up part TS(t), and the boundary between the deceleration part and the speed-up part is the speed demarcation point; When a left-turning vehicle travels from the third entrance road to the fourth exit road, the speed cutoff point is: (1); In formula (1), A1, A2, A3 and A4 are constants, and t is the time; Step S33, calculating the deceleration portion JS(t) according to the estimated minimum speed; wherein the initial minimum speed XSDc is calculated by the left turn arc value ZHD of the target traffic road, the lateral distance HJL between the left-turning vehicle and the road edge, and the turning radius ZWB of the left-turning vehicle when turning; XSDc=C1×ZHD+C2×HJL+C3×ZWB (2); Wherein, the initial minimum speed is the estimated speed of the left-turn lane reaching the speed demarcation point, and C1, C2, and C3 are calculated by maximum likelihood estimation based on the driving conditions of historical left-turning vehicles when making left turns.
[0008] Furthermore, the step S3 also includes the following sub-steps: Step S34, when the real-time speed of the left-turning vehicle becomes the minimum, JS(t) and tmin are estimated, tmin is the time when the left-turning vehicle reaches the speed demarcation point, and the following equation can be obtained from the constraint condition of the speed demarcation point: (3); (4); In formula (4), LJC refers to the path length from the current vehicle position to the speed demarcation point, A4 is equal to the speed at the current moment, and A3 is equal to the acceleration at the current moment; Step S35, combine equations (1), (3), and (4) to solve tmin, A1, and A2: (5); (6); (7).
[0009] Furthermore, the crosswalk video data is the pedestrian passage time corresponding to multiple pedestrians passing through the crosswalk during each green light period.
[0010] Furthermore, the step S4 includes the following sub-steps: Step S41, reading the crosswalk video data of the target traffic road, and obtaining the pedestrian passage time XTSn corresponding to multiple pedestrians passing through the crosswalk during each green light period, where n is the number of multiple pedestrians n=1, 2, ..., z, and z is a positive integer; Step S42, adding up the pedestrian travel time of multiple pedestrians and taking the average value to obtain the average pedestrian travel time XPS of the pedestrians during the green light period; Step S43, the pedestrian passage variance XFC corresponding to the pedestrian during the green light period is calculated by a formula, and the specific formula is: ; Step S44, obtaining a travel time threshold by adding the pedestrian average travel time to the pedestrian travel variance; Step S45, obtaining the left-turn speed model of the left-turning vehicle corresponding to the target road, obtaining the time tmin corresponding to the left-turning vehicle reaching the speed demarcation point and the time t0 when the left-turning vehicle starts to turn left, and calculating the vehicle passage time of the left-turning vehicle by tmin-t0; Step S46, comparing the vehicle's travel time with the travel time threshold; If the vehicle's travel time is less than the travel time threshold, the waiting time for the left-turning vehicle during the green light period is obtained by subtracting the vehicle's travel time from the travel time threshold. If the vehicle's travel time is greater than or equal to the travel time threshold, the waiting time for the left-turning vehicle during the green light period is zero, which is obtained by subtracting the vehicle's travel time from the travel time threshold. Step S47, similarly calculate the waiting time of vehicles turning left during multiple green lights corresponding to the crosswalk video data, and select the mode of the waiting time as the comprehensive pedestrian congestion value ZXY of the target traffic road.
[0011] Furthermore, the step S5 includes the following sub-steps: Step S51, obtaining the historical congestion value LSY and the comprehensive pedestrian congestion value ZXY corresponding to the target traffic road; Step S52, calculating the road congestion level value DYD of the target traffic road by a formula, the specific calculation formula is: DYD=B1×LSY+B2×ZXY; B1 and B2 are proportional coefficients, 0<B2<B1<1; Step S53, comparing the road congestion level value with the congestion level threshold, determining the road congestion level of the target traffic road as the first road congestion level, the second road congestion level or the third road congestion level; Step S54, sending the road congestion level of the target traffic road to the user terminal, so as to obtain the congestion situation of the target traffic road through the user terminal.
[0012] Furthermore, the probability of road congestion occurring on the target traffic road corresponding to the third road congestion level is greater than the probability of road congestion occurring on the target traffic road corresponding to the second road congestion level, and the probability of road congestion occurring on the target traffic road corresponding to the second road congestion level is greater than the probability of road congestion occurring on the target traffic road corresponding to the first road congestion level.
[0013] In the second aspect, in order to achieve the above-mentioned purpose, the present invention also adopts the following technical solution: a traffic congestion prediction system based on artificial intelligence, the system comprising: A database, used to store historical traffic data corresponding to the target traffic road; A historical analysis module is used to analyze the historical traffic data of the target traffic road to obtain the historical congestion value of the target traffic road; A model building module, used to build a left-turn speed model of left-turning vehicles corresponding to the target traffic road based on historical traffic data corresponding to the target traffic road; Data acquisition module, used for pedestrian crossing video data of target traffic road; A data analysis module is used to analyze the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road; A comprehensive analysis module is used to analyze the road congestion situation of the target traffic road by combining the historical congestion value and the comprehensive pedestrian congestion value, and obtain the road congestion level corresponding to the target traffic road; The user terminal is used to check the road congestion level of the target traffic road.
[0014] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The present invention firstly obtains historical traffic data corresponding to a target traffic road, analyzes the historical traffic data of the target traffic road based on the historical traffic data on the one hand, and obtains a historical congestion value of the target traffic road, and on the other hand obtains pedestrian crossing video data of the target traffic road, and simultaneously constructs a left-turn speed model of left-turning vehicles corresponding to the target traffic road, and obtains the vehicle travel time of the left-turning vehicles by using the left-turn speed model analysis, and analyzes the pedestrian crossing video data in combination with the vehicle travel time to obtain a comprehensive pedestrian congestion value of the target traffic road, and finally obtains the road congestion level corresponding to the target traffic road by using the historical congestion value and the comprehensive pedestrian congestion value analysis, and the present invention combines two dimensional data of road historical traffic conditions and personnel traffic conditions to realize accurate prediction of traffic road congestion conditions. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.
[0016] Figure 1 The present invention is a flow chart of the method.
[0017] Figure 2 It is a schematic diagram of the target traffic road in the present invention.
[0018] Figure 3 It is a left turn schematic diagram of a left-turning vehicle in the present invention.
[0019] Figure 4 It is a schematic diagram of parameters of a left-turning vehicle during a left-turning process in the present invention.
[0020] Figure 5 It is a schematic diagram of the deceleration part JS (t) in the present invention.
[0021] Figure 6 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION
[0022] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0023] Embodiment 1: See also Figure 1-Figure 5 As shown, the technical solution provided by the present invention is: a traffic road congestion prediction method based on artificial intelligence, the method is applied to a traffic road where there is a crossroad or a T-junction, and the corresponding traffic road is defined as a target traffic road, the method includes: Step S1, obtaining historical traffic data corresponding to the target traffic road; the historical traffic data specifically includes the road daily traffic flow of the target traffic road in the same month as the current time in the previous year, and the historical average vehicle speed and historical average vehicle spacing of the target traffic road in the same month as the current time in the previous year.
[0024] Step S2, analyzing the historical traffic data of the target traffic road to obtain the historical congestion value of the target traffic road; Step S21, acquiring historical traffic data of the target traffic road, obtaining the historical average vehicle speed LSD, the historical average vehicle spacing LCJ and the road daily traffic flow RLL of the target traffic road in the same month as the current time in the previous year; Step S22, obtaining the maximum traffic volume TXL of the target traffic road; Among them, the maximum road traffic volume is the basic attribute of the target traffic road, that is, the theoretical maximum vehicle traffic volume of the target traffic road, which is only related to the road width and the number of lanes of the target traffic road; Step S23, the historical congestion value LSY of the target traffic road is calculated by a formula, and the specific formula is: LSY=1 / LSD+e / LCJ+RLL / TXL; where e is a natural constant.
[0025] Step S3, see Figure 3-Figure 5 As shown, a left-turn speed model of left-turn vehicles corresponding to the target traffic road is constructed based on the historical traffic data corresponding to the target traffic road; In this embodiment, step S3 specifically includes the following sub-steps: Step S31, obtaining the left turn arc value corresponding to the left turn operation performed by the left-turning vehicle on the target traffic road on the same day, the lateral distance between the left-turning vehicle and the road edge, and the turning radius of the left-turning vehicle when turning; like Figure 4As shown, a vertical line is drawn from the tangent of the left-turning vehicle at the current moment, and then a vertical line is drawn from the tangent at the speed dividing point. The intersection of the two vertical lines is the center of the left-turn arc, and the corresponding angle value is the left-turn arc value; the turning radius is the distance from the left-turning vehicle to the center of the circle; Step S32, as Figure 5 The coordinate diagram corresponding to the left-turn speed model is shown, and the left-turn speed model of the left-turn vehicle is constructed. The point where the corresponding speed of the left-turn vehicle is the smallest during the turning process is defined as the speed demarcation point; the left-turn speed model is divided into two parts, the deceleration part JS (t) and the acceleration part TS (t), and the boundary between the deceleration part and the acceleration part is the speed demarcation point. In this embodiment, only the deceleration part JS (t) is considered; like Figure 3 As shown, when the left-turning vehicle travels from the third entrance road to the fourth exit road, the speed demarcation point is as shown in the figure; (1); In formula (1), A1, A2, A3 and A4 are constants, and t is the time; Considering the interaction between left-turning vehicles and pedestrians, it is assumed that the speed of the left-turning vehicle becomes the minimum when it arrives around the crosswalk. In this embodiment, the speed of the left-turning vehicle is considered to be the minimum when the front of the left-turning vehicle is one parking space away from the crosswalk. Step S33, as Figure 3 As shown, the deceleration part JS(t) is calculated based on the estimated minimum speed; wherein the initial minimum speed XSDc is calculated by the left turn arc value ZHD of the target traffic road, the lateral distance HJL between the left-turning vehicle and the road edge, and the turning radius ZWB of the left-turning vehicle when turning; XSDc=C1×ZHD+C2×HJL+C3×ZWB (2); In formula (2), the initial minimum speed is the estimated speed of the left-turn lane reaching the speed demarcation point, and C1, C2, and C3 are calculated by maximum likelihood estimation based on the driving conditions of historical left-turning vehicles when making left turns; Step S34, when the real-time speed of the left-turning vehicle becomes the minimum, JS(t) and tmin are estimated, tmin is the time when the left-turning vehicle reaches the speed demarcation point, and the following equation can be obtained from the constraint condition of the speed demarcation point: (3); (4); In formula (4), LJC refers to the path length from the current vehicle position to the speed demarcation point, A4 is equal to the speed at the current moment, and A3 is equal to the acceleration at the current moment. In actual calculation, in order to consider the influence of errors, A4 is located between [XSDc-3, XSDc+3] in this embodiment. Step S35, combine formulas (1), (3), and (4) to obtain tmin, A1, and A2: (5); (6); (7).
[0026] Step S4, obtaining the pedestrian crossing video data of the target traffic road, and analyzing the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road; The pedestrian crossing video data specifically refers to the relevant videos of pedestrians crossing the pedestrian crossing collected by the surveillance camera, and the pedestrian crossing video data refers to the pedestrian passage time corresponding to multiple pedestrians passing through the pedestrian crossing during each green light period; In this embodiment, step S4 specifically includes the following sub-steps: Step S41, reading the crosswalk video data of the target traffic road, and obtaining the pedestrian passage time XTSn corresponding to multiple pedestrians passing through the crosswalk during each green light period, where n is the number of multiple pedestrians n=1, 2, ..., z, and z is a positive integer; Step S42, adding up the pedestrian travel time of multiple pedestrians and taking the average value to obtain the average pedestrian travel time XPS of the pedestrians during the green light period; Step S43, the pedestrian passage variance XFC corresponding to the pedestrian during the green light period is calculated by a formula, and the specific formula is: ; Step S44, obtaining a travel time threshold by adding the pedestrian average travel time to the pedestrian travel variance; Step S45, obtaining the left-turn speed model of the left-turning vehicle corresponding to the target road, obtaining the time tmin corresponding to the left-turning vehicle reaching the speed demarcation point and the time t0 when the left-turning vehicle starts to turn left, and calculating the vehicle passage time of the left-turning vehicle by tmin-t0; Step S46, comparing the vehicle's travel time with the travel time threshold; If the vehicle's travel time is less than the travel time threshold, the waiting time for the left-turning vehicle during the green light period is obtained by subtracting the vehicle's travel time from the travel time threshold. If the vehicle's travel time is greater than or equal to the travel time threshold, the waiting time for the left-turning vehicle during the green light period is zero, which is obtained by subtracting the vehicle's travel time from the travel time threshold. Step S47, similarly calculate the waiting time of vehicles turning left during multiple green lights corresponding to the crosswalk video data, and select the mode of the waiting time as the comprehensive pedestrian congestion value ZXY of the target traffic road.
[0027] Step S5, combining the historical congestion value of the target traffic road with the comprehensive pedestrian congestion value to obtain the road congestion level corresponding to the target traffic road; In this embodiment, step S5 includes the following sub-steps: Step S51, obtaining the historical congestion value LSY and the comprehensive pedestrian congestion value ZXY corresponding to the target traffic road; Step S52, calculating the road congestion level value DYD of the target traffic road by a formula, the specific calculation formula is: DYD=B1×LSY+B2×ZXY; B1 and B2 are proportional coefficients, 0<B2<B1<1; Step S53, comparing the road congestion level value with the congestion level threshold, if the road congestion level value is less than or equal to the first congestion level threshold, recording the road congestion level of the target traffic road as the first road congestion level; If the road congestion level value is greater than the first congestion level threshold, and the road congestion level value is less than or equal to the second congestion level threshold, the road congestion level of the target traffic road is recorded as the second road congestion level; If the road congestion level value is greater than the second congestion level threshold, the road congestion level of the target traffic road is recorded as the third road congestion level; The second congestion level threshold is greater than the first congestion level threshold; the probability of road congestion occurring on the target traffic road corresponding to the third road congestion level is greater than the probability of road congestion occurring on the target traffic road corresponding to the second road congestion level, and the probability of road congestion occurring on the target traffic road corresponding to the second road congestion level is greater than the probability of road congestion occurring on the target traffic road corresponding to the first road congestion level; Step S54, sending the road congestion level of the target traffic road to the user terminal, so as to obtain the congestion situation of the target traffic road through the user terminal.
[0028] The present invention adopts the method of the present embodiment, firstly, obtaining historical traffic data corresponding to the target traffic road, on the one hand, analyzing the historical traffic data of the target traffic road based on the historical traffic data to obtain the historical congestion value of the target traffic road, on the other hand, obtaining the pedestrian crossing video data of the target traffic road, and at the same time constructing a left turn speed model of left-turning vehicles corresponding to the target traffic road, using the left turn speed model to analyze the vehicle travel time of the left-turning vehicles, combining the vehicle travel time to analyze the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road, and finally using the historical congestion value and the comprehensive pedestrian congestion value to analyze the road congestion level corresponding to the target traffic road. The present embodiment combines the two dimensional data of road historical traffic conditions and personnel traffic conditions to achieve accurate prediction of traffic road congestion conditions.
[0029] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.
[0030] Embodiment 2: Figure 6 As shown, based on another concept of the same invention, a traffic road congestion prediction system based on artificial intelligence is now proposed, and the system includes: A database, used to store historical traffic data corresponding to the target traffic road; A historical analysis module is used to analyze the historical traffic data of the target traffic road to obtain the historical congestion value of the target traffic road; A model building module, used to build a left-turn speed model of left-turning vehicles corresponding to the target traffic road based on historical traffic data corresponding to the target traffic road; Data acquisition module, used for pedestrian crossing video data of target traffic road; A data analysis module is used to analyze the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road; A comprehensive analysis module is used to analyze the road congestion situation of the target traffic road by combining the historical congestion value and the comprehensive pedestrian congestion value, and obtain the road congestion level corresponding to the target traffic road; The user terminal is used to check the road congestion level of the target traffic road.
[0031] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. A traffic congestion prediction method based on artificial intelligence, characterized in that: The traffic congestion prediction method is as follows: Step S1, obtaining historical traffic data corresponding to the target traffic road; Step S2, analyzing the historical traffic data of the target traffic road to obtain the historical congestion value of the target traffic road; Step S3, constructing a left-turn speed model for left-turning vehicles corresponding to the target traffic road according to historical traffic data corresponding to the target traffic road; Step S4, obtaining the pedestrian crossing video data of the target traffic road, and analyzing the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road; Step S5, combining the historical congestion value of the target traffic road with the comprehensive pedestrian congestion value to obtain the road congestion level corresponding to the target traffic road.
2. The method for predicting traffic congestion based on artificial intelligence according to claim 1, characterized in that: The historical traffic data includes the daily traffic volume of the target road in the same month as the current time in the previous year, and the historical average vehicle speed and historical average vehicle spacing of the target traffic road in the same month as the current time in the previous year.
3. The method for predicting traffic congestion based on artificial intelligence according to claim 2, characterized in that: The step S2 includes the following sub-steps: Step S21, acquiring historical traffic data of the target traffic road, obtaining the historical average vehicle speed LSD, the historical average vehicle spacing LCJ and the road daily traffic flow RLL of the target traffic road in the same month as the current time in the previous year; Step S22, obtaining the maximum traffic volume TXL of the target traffic road; Step S23, calculating the historical congestion value LSY of the target traffic road by the formula LSY=1 / LSD+e / LCJ+RLL / TXL; wherein e is a natural constant.
4. The method for predicting traffic congestion based on artificial intelligence according to claim 1, characterized in that: The step S3 includes the following sub-steps: Step S31, obtaining the left turn arc value corresponding to the left turn operation performed by the left-turning vehicle on the target traffic road on the same day, the lateral distance between the left-turning vehicle and the road edge, and the turning radius of the left-turning vehicle when turning; Step S32, constructing a left-turn speed model for the left-turn vehicle, defining the point where the left-turn vehicle has the minimum speed during the turning process as the speed demarcation point; the left-turn speed model is divided into two parts, a deceleration part JS(t) and a speed-up part TS(t), and the boundary between the deceleration part and the speed-up part is the speed demarcation point; When a left-turning vehicle travels from the third entrance road to the fourth exit road, the speed cutoff point is: （1）； In formula (1), A1, A2, A3 and A4 are constants, and t is the time; Step S33, calculating the deceleration portion JS(t) according to the estimated minimum speed; wherein the initial minimum speed XSDc is calculated by the left turn arc value ZHD of the target traffic road, the lateral distance HJL between the left-turning vehicle and the road edge, and the turning radius ZWB of the left-turning vehicle when turning; XSDc=C1×ZHD+C2×HJL+C3×ZWB (2); Wherein, the initial minimum speed is the estimated speed of the left-turn lane reaching the speed demarcation point, and C1, C2, and C3 are calculated by maximum likelihood estimation based on the driving conditions of historical left-turning vehicles when making left turns.
5. The method for predicting traffic congestion based on artificial intelligence according to claim 4, characterized in that: The step S3 further comprises the following sub-steps: Step S34, when the real-time speed of the left-turning vehicle becomes the minimum, JS(t) and tmin are estimated, tmin is the time when the left-turning vehicle reaches the speed demarcation point, and the following equation can be obtained from the constraint condition of the speed demarcation point: （3）； （4）； In formula (4), LJC refers to the path length from the current vehicle position to the speed demarcation point, A4 is equal to the speed at the current moment, and A3 is equal to the acceleration at the current moment; Step S35, combine equations (1), (3), and (4) to solve tmin, A1, and A2: （5）； （6）； （7）。 6. The method for predicting traffic congestion based on artificial intelligence according to claim 1, characterized in that: The pedestrian crossing video data is the pedestrian passage time corresponding to multiple pedestrians passing through the pedestrian crossing during each green light period.
7. The method for predicting traffic congestion based on artificial intelligence according to claim 6, characterized in that: The step S4 includes the following sub-steps: Step S41, reading the crosswalk video data of the target traffic road, and obtaining the pedestrian passage time XTSn corresponding to multiple pedestrians passing through the crosswalk during each green light period, where n is the number of multiple pedestrians n=1, 2, ..., z, and z is a positive integer; Step S42, adding up the pedestrian travel time of multiple pedestrians and taking the average value to obtain the average pedestrian travel time XPS of the pedestrians during the green light period; Step S43, the pedestrian passage variance XFC corresponding to the pedestrian during the green light period is calculated by a formula, and the specific formula is: ； Step S44, obtaining a travel time threshold by adding the pedestrian average travel time to the pedestrian travel variance; Step S45, obtaining the left-turn speed model of the left-turning vehicle corresponding to the target road, obtaining the time tmin corresponding to the left-turning vehicle reaching the speed demarcation point and the time t0 when the left-turning vehicle starts to turn left, and calculating the vehicle passage time of the left-turning vehicle by tmin-t0; Step S46, comparing the vehicle's travel time with the travel time threshold; If the vehicle's travel time is less than the travel time threshold, the waiting time for the left-turning vehicle during the green light period is obtained by subtracting the vehicle's travel time from the travel time threshold. If the vehicle's travel time is greater than or equal to the travel time threshold, the waiting time for the left-turning vehicle during the green light period is zero, which is obtained by subtracting the vehicle's travel time from the travel time threshold. Step S47, similarly calculate the waiting time of vehicles turning left during multiple green lights corresponding to the crosswalk video data, and select the mode of the waiting time as the comprehensive pedestrian congestion value ZXY of the target traffic road.
8. The method for predicting traffic congestion based on artificial intelligence according to claim 1, characterized in that: The step S5 comprises the following sub-steps: Step S51, obtaining the historical congestion value LSY and the comprehensive pedestrian congestion value ZXY corresponding to the target traffic road; Step S52, calculating the road congestion level value DYD of the target traffic road by a formula, the specific calculation formula is: DYD=B1×LSY+B2×ZXY; B1 and B2 are proportional coefficients, 0<B2<B1<1; Step S53, comparing the road congestion level value with the congestion level threshold, determining the road congestion level of the target traffic road as the first road congestion level, the second road congestion level or the third road congestion level; Step S54, sending the road congestion level of the target traffic road to the user terminal, so as to obtain the congestion situation of the target traffic road through the user terminal.
9. The method for predicting traffic congestion based on artificial intelligence according to claim 8, characterized in that: The probability of road congestion occurring on the target traffic road corresponding to the third road congestion level is greater than the probability of road congestion occurring on the target traffic road corresponding to the second road congestion level, and the probability of road congestion occurring on the target traffic road corresponding to the second road congestion level is greater than the probability of road congestion occurring on the target traffic road corresponding to the first road congestion level.
10. The traffic congestion prediction system based on artificial intelligence is characterized by: Using the traffic congestion prediction method based on artificial intelligence according to any one of claims 1 to 9, the system comprises: A database, used to store historical traffic data corresponding to the target traffic road; A historical analysis module is used to analyze the historical traffic data of the target traffic road to obtain the historical congestion value of the target traffic road; A model building module, used to build a left-turn speed model of left-turning vehicles corresponding to the target traffic road based on historical traffic data corresponding to the target traffic road; Data acquisition module, used for pedestrian crossing video data of target traffic road; A data analysis module is used to analyze the pedestrian crossing video data to obtain the comprehensive pedestrian congestion value of the target traffic road; A comprehensive analysis module is used to analyze the road congestion situation of the target traffic road by combining the historical congestion value and the comprehensive pedestrian congestion value, and obtain the road congestion level corresponding to the target traffic road; The user terminal is used to check the road congestion level of the target traffic road.