A lane-level congestion prediction method for urban intelligent network hybrid traffic flow

Abstract

The application discloses a lane-level congestion prediction method for urban intelligent network connection mixed traffic flow. The method comprises six steps, namely, acquiring the instantaneous displacement of intelligent network connection vehicles and the instantaneous speed of intelligent network connection vehicles, measuring the instantaneous displacement of non-intelligent network connection vehicles and the instantaneous speed of non-intelligent network connection vehicles, calculating the average speed of mixed traffic flow on a target lane, searching for characteristic points of the average speed time sequence of mixed traffic flow on the target lane, calculating the total disturbance of mixed traffic flow on the target lane and searching for characteristic points of the total disturbance time sequence of mixed traffic flow, and predicting the ending time of single congestion change of the target lane and the average speed of mixed traffic flow at the time. The method models and expresses the influence of the change of the micro-vehicle motion state of each lane of a road section on the speed and congestion evolution of the target lane, takes the collected real-time traffic flow data as the model input, and outputs the prediction results of the congestion trend and the traffic flow speed, so that the purpose of preventing and relieving congestion is achieved.

Description

Technical Field

[0001] This invention relates to the field of lane-level prediction of urban traffic flow, and specifically to a lane-level congestion prediction method for intelligent connected mixed traffic flow in cities. Background Technology

[0002] With the rapid development of intelligent connected and autonomous driving technologies, mixed traffic flows composed of conventional vehicles and intelligent connected vehicles will exist for a long time to come. The multi-source perception fusion of C-V2X onboard terminal equipment and intelligent roadside equipment provides support for the accurate and real-time acquisition of vehicle operation information. Based on the comprehensive collection of urban mixed traffic flow operation data, traffic flow speed prediction and determination of future trends in traffic operation status based on historical traffic information and real-time conditions are of great significance for achieving precise control of urban traffic conditions, timely prevention and rapid alleviation of congestion, and improvement of road service levels and operational efficiency.

[0003] Existing congestion trend analysis and lane-level traffic flow prediction technologies mostly employ time series analysis, probabilistic statistical methods, and machine learning algorithms. They capture the uncertainty and complex characteristics of traffic flow parameter time-series changes to model traffic conditions and predict traffic flow in specific scenarios. For example, patent 202010262372.7 uses LSTM to predict the correlation between different lane sections of the same road segment; patent 202110114317.8 improves traffic flow prediction accuracy to the lane level by building a neural network with an embedded spatiotemporal attention module; patent 202410397699.3 models and predicts future conflict risks for each lane based on machine learning methods; and patent 201710065260.0 introduces a vehicle risk threshold, considering its impact on driving behavior, and achieves prediction of lane-level queue length and average speed. However, existing technologies neglect the lag effect between the spatiotemporal distribution of microscopic vehicle motion differences and changes in traffic conditions, and machine learning-based technologies require a large amount of historical data for traffic flow parameter feature extraction and model training. In addition to predicting the average speed of traffic flow, it is also essential to predict key time points for changes in congestion trends in practical applications.

[0004] Based on the above background, there is an urgent need to design a lane-level congestion prediction method for intelligent connected mixed traffic flow. This method should model and express the impact of changes in the microscopic vehicle motion state of each lane on the target lane speed and congestion evolution process. The collected real-time traffic flow data should be used as the model input, and the output should be the prediction results of congestion trend and average traffic flow speed, thereby achieving the goal of preventing and alleviating congestion. Summary of the Invention

[0005] The technical problem to be solved by this invention is to address the shortcomings of existing technologies by proposing a lane-level congestion prediction method for urban intelligent connected mixed traffic flow. This method can accurately predict the start time of congestion changes and the average speed of the mixed traffic flow at that time, providing theoretical and technical support for lane-level prediction and traffic flow organization and control in urban traffic flow.

[0006] The present invention is achieved using the following technical solution, as described below:

[0007] Step 1: Obtain the instantaneous displacement and instantaneous speed of the intelligent connected vehicle;

[0008] A coordinate system is established with the intersection of the stop line and the center line of the road at the downstream intersection adjacent to the data collection area as the origin, the driving direction for which congestion prediction is to be performed as the positive X-axis, and the positive Y-axis as the direction of rotation 90 degrees counterclockwise along the positive X-axis.

[0009] Data collected via vehicle-mounted or roadside equipment and Let t be the time of data acquisition, a be the ID of the intelligent connected vehicle that passes through the data collection area at time t, where a = 1, 2, ..., m(t), and m(t) be the number of intelligent connected vehicles that pass through the data collection area at time t. The instantaneous displacement of the intelligent connected vehicle a at time t consists of two parts, wherein... Let be the x-coordinate of the instantaneous displacement of the intelligent connected vehicle a at time t. Let be the ordinate of the instantaneous displacement of the intelligent connected vehicle a at time t. Let be the instantaneous speed of intelligent connected vehicle a at time t;

[0010] Step 2: Calculate the instantaneous displacement and instantaneous speed of the non-intelligent connected vehicle.

[0011] Let b be the ID of the non-intelligent connected vehicle that passes through the data collection area at time t, where b = 1, 2, ..., n(t), and n(t) be the number of non-intelligent connected vehicles that pass through the data collection area at time t. The instantaneous displacement of the non-intelligent connected vehicle b at time t consists of two parts, wherein... Let x be the abscissa of the instantaneous displacement of the non-intelligent connected vehicle b at time t. Let be the ordinate of the instantaneous displacement of the non-intelligent connected vehicle b at time t. Let be the instantaneous speed of the non-intelligent connected vehicle b at time t;

[0012] For non-intelligent connected vehicles within the perception range of intelligent connected vehicles, data is collected through onboard or roadside equipment. and For non-intelligent connected vehicles that are outside the perception range of intelligent connected vehicles, the distance is estimated based on the instantaneous displacement and instantaneous speed of the nearest intelligent connected vehicles in front and behind. and

[0013]

[0014] in, To and Functions related to q(t) and Q(t), To and Functions related to q(t) and Q(t), for Functions related to q(t), Q(t), q′(t), and Q′(t); at time t, the intelligent connected vehicles adjacent to the non-intelligent connected vehicle numbered b are numbered a, a, and a, respectively. f With a r The non-intelligent connected vehicle numbered b is adjacent to intelligent connected vehicles numbered a on both sides. f ′ and a r At time t, along the X-axis, the intelligent connected vehicle a... f With intelligent connected vehicles a r The number of non-intelligent connected vehicles between them is Q(t), and non-intelligent connected vehicle b is the number of intelligent connected vehicles a. r The nearest non-intelligent connected vehicle q(t); at time t, along the Y-axis, intelligent connected vehicle a f With intelligent connected vehicles a r The number of non-intelligent connected vehicles between points A and B is Q′(t), and non-intelligent connected vehicle b is connected to intelligent connected vehicle a. r The q'(t)th non-intelligent connected vehicle in the nearest neighbor; For intelligent connected vehicles at time t r instantaneous speed, For intelligent connected vehicles at time t r The instantaneous displacement x-coordinate For intelligent connected vehicles at time t r The instantaneous displacement ordinate; For intelligent connected vehicles at time t f instantaneous speed, For intelligent connected vehicles at time t f The instantaneous displacement x-coordinate For intelligent connected vehicles at time t f The instantaneous displacement ordinate, For intelligent connected vehicles at time t r The instantaneous velocity of ′ For intelligent connected vehicles at time tf The instantaneous velocity of ′;

[0015] Step 3: Calculate the average speed of mixed traffic flow on the target lane.

[0016] Let be the average of the instantaneous speeds of all intelligent connected vehicles and all non-intelligent connected vehicles in the target lane at time t:

[0017]

[0018] A(t) is the number of mixed traffic vehicles in the target lane at time t, which is the sum of the number of all intelligent connected vehicles m′(t) and the number of all non-intelligent connected vehicles n′(t) in the target lane. A(t) = m′(t) + n′(t).

[0019] Step 4: Search for time series feature points of average speed of mixed traffic flow on the target lane;

[0020] There are four categories of time series characteristic points for the average speed of mixed traffic flow, defined as the congestion initiation point T1′(t), congestion initiation point T2′(t), congestion dissipation point T3′(t), and congestion dissipation point T4′(t). During the search, these four categories of time series characteristic points for the average speed of mixed traffic flow conform to the following mathematical characteristics:

[0021] (1) The starting point of congestion formation, T1′(t), is... The moment when the high-speed fluctuation turns into a monotonically decreasing maximum, where the range of the high-speed fluctuation is... Not less than 40km / h;

[0022] (2) The congestion formation endpoint T2′(t) is: The time corresponding to the minimum point where the monotonically decreasing trend turns into a low-speed fluctuation, where the range of the low-speed fluctuation is... Not exceeding 30km / h;

[0023] (3) The congestion dissipation starting point T3′(t) is The moment when the low-speed fluctuation turns into a monotonically increasing minimum value;

[0024] (4) The congestion dissipation endpoint T4′(t) is The moment when the monotonically increasing trend turns into a high-speed fluctuation at its maximum value;

[0025] Step 5: Calculate the total turbulence of mixed traffic flow on the target lane and search for time series feature points of the total turbulence of mixed traffic flow;

[0026] The data collection area is L in length, the number of lanes in the driving direction for which congestion prediction is to be performed is Z, and the target lane is numbered z0. The data collection area is divided into Z x K equal-sized road grids, and the total turbulence E of the mixed traffic flow on the target lane at time t is calculated. S (z0,t):

[0027]

[0028] Where z0 is the number of the road grid along the Y-axis, z0 = 1, 2, ..., Z, k is the number of the road grid along the X-axis, k = 1, 2, ..., K, K is the total number of road grids on a single lane, E C (z0,k,t) represents the mixed traffic flow turbulence on the road grid (z0,k) at time t, E S (z0,t) represents the total turbulence of mixed traffic flow on the target lane at time t, where i is the vehicle number. Let be the distance from the geometric center of vehicle i at time t to the geometric center of the road grid numbered (z0,k). v represents the average vehicle spacing of mixed traffic flow in the data collection area at time t. i (t)cosδ i With v j (t)cosδ j Vehicles i and j at time t are respectively... The component of velocity in direction, v i (t)sinδ i With v j (t)sinδ j Let i and j be perpendicular to each other at time t. Component of velocity in direction, For time t, vehicle i and vehicle j are along The displacement difference in direction, where G represents the vehicle type, G∈{I,N}. At time t, vehicle i is perpendicular to vehicle j. Displacement difference in direction;

[0029] There are four categories of time series feature points for total turbulence in mixed traffic flow: the starting point T1(t) corresponding to the increase in total turbulence caused by congestion, the ending point T2(t) corresponding to the increase in total turbulence caused by congestion, the starting point T3(t) corresponding to the increase in total turbulence caused by congestion dissipation, and the ending point T4(t) corresponding to the increase in total turbulence caused by congestion dissipation. During the search, these four categories of time series feature points for total turbulence in mixed traffic flow conform to the following mathematical characteristics:

[0030] (1) The starting point of the increase in total turbulence of mixed traffic flow corresponding to congestion is T1(t), which is the congestion formation stage E. S(z0,t) represents the moment corresponding to the turning point where the low-value fluctuations begin to shift towards a significant increase, where the range of the low-value fluctuations is... No more than 300;

[0031] (2) The endpoint T2(t) of the increase in total turbulence of the mixed traffic flow corresponding to the congestion is the congestion formation stage E. S (z0,t) ends its significant increase and turns to the moment corresponding to the turning point of low-value fluctuation;

[0032] (3) The starting point of the increase in total turbulence of mixed traffic flow corresponding to congestion dissipation, T3(t), is the congestion dissipation stage E. S (z0,t) represents the moment when the low-value fluctuations begin to turn into a significant increase.

[0033] (4) The endpoint T4(t) of the total turbulence increase in the mixed traffic flow corresponding to congestion dissipation is the congestion dissipation stage E. S (z0,t) ends its significant increase and turns to the moment corresponding to the turning point of low-value fluctuation;

[0034] Step 6: Predict the end time T of a single congestion change in the target lane. a ′ and T a Average speed of mixed traffic flow at time ′

[0035] E S (z0,t) The time T at which the change begins p Including T1(t) and T3(t); E S (z0,t) The time T at which the change ends a Including T2(t) and T4(t); the starting time T of the target lane congestion change. p Including T1′(t) and T3′(t); the end time T of the congestion change in the target lane. a This includes T2′(t) and T4′(t);

[0036] Average speed of mixed traffic flow at the start of congestion change Including the average speed of mixed traffic flow at the point where congestion begins Average speed of mixed traffic flow corresponding to the congestion dissipation point Average speed of mixed traffic flow at the end of congestion change Including the average speed of mixed traffic flow at the endpoint of congestion formation Average speed of mixed traffic flow corresponding to the congestion dissipation endpoint

[0037] By observing T p T p ′、T a , And calculate E at each time point S Solve for (z,t) T2′(t) and T4′(t):

[0038]

[0039] Where c w With c w ′ represents the polynomial coefficients, w = 0, 1, 2, 3, 4; ΔE S (z0,t) represents E at time t. S The change in (z0,t) compared to the previous moment, ΔE S (z0,t)=E S (z0,t)-E S (z0,t-1); s is the average velocity of the mixed traffic flow in the monotonic interval [T]. p ′,T a The slope on ′].

[0040] Furthermore, the data collection area in step one is no less than 50 meters in length and the width is equal to all Z lanes in the driving direction for which congestion prediction is to be performed. The time interval between the data acquisition time t and the previous time t-1 does not exceed 2 seconds.

[0041] Furthermore, in step two, according to Calculation of q(t), Q(t), q′(t) and Q′(t) and The specific method is as follows:

[0042]

[0043] Furthermore, in step six, s is approximately equal to the average speed of the mixed traffic flow in the monotonic interval [T]. p ′,T a The slope on ], i.e.

[0044] Compared with existing technologies, the advantages of the lane-level congestion prediction method for urban intelligent connected mixed traffic flow described in this invention are:

[0045] (1) The method considers the impact of the spatiotemporal transmission of the differences in motion states between micro vehicles on the macro traffic flow congestion development process. It establishes a polynomial relationship model between the average speed time series of single-lane mixed traffic flow and the total turbulence time series of single-lane mixed traffic flow. The model input is the real-time trajectory information of intelligent connected vehicles collected and measured. It does not require a large amount of historical traffic data for model training or calibration. It realizes the prediction of the end time of congestion formation and the end time of congestion dissipation, and outputs the predicted value of the average speed of traffic flow at the end time of congestion change. It improves the prediction accuracy to the lane level and provides decision-making basis and technical support for vehicle path guidance, traffic flow organization and control and congestion prevention and evacuation.

[0046] (2) The method collects the instantaneous speed and instantaneous displacement of intelligent connected vehicles and non-intelligent connected vehicles within the perception range of intelligent connected vehicles through road testing equipment and on-board equipment of intelligent connected vehicles (including but not limited to cameras, millimeter-wave radar, lidar, ultrasonic sensors, etc.); for non-intelligent connected vehicles outside the perception range of intelligent connected vehicles, the instantaneous displacement and instantaneous speed of non-intelligent connected vehicles are estimated based on their relative positional relationship with the intelligent connected vehicles in front and behind, thereby realizing the real-time collection of mixed traffic flow operation parameters without having to meet the requirement of high road detector coverage, thus reducing the data acquisition cost. Attached Figure Description

[0047] Figure 1 This is a flowchart of the lane-level congestion prediction method for urban intelligent connected mixed traffic flow as described in this invention.

[0048] Figure 2 This is a schematic diagram of the coordinate system described in this invention;

[0049] Figure 3 This is a schematic diagram illustrating the calculation of the instantaneous displacement and instantaneous speed of a non-intelligent connected vehicle as described in this invention.

[0050] Figure 4 This is a schematic diagram illustrating the calculation of total turbulence of mixed traffic flow on the target lane as described in this invention;

[0051] Figure 5 This is a schematic diagram of the time series feature points of the average speed of the mixed traffic flow on the target lane and the time series feature points of the average speed of the mixed traffic flow on the target lane as described in this invention. Detailed Implementation

[0052] The detailed content of the present invention and the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0053] See Figure 1 The lane-level congestion prediction method for urban intelligent connected mixed traffic flow described in this invention consists of the following steps:

[0054] Step 1: Obtain the instantaneous displacement and instantaneous speed of the intelligent connected vehicle;

[0055] Step 2: Calculate the instantaneous displacement and instantaneous speed of the non-intelligent connected vehicle.

[0056] Step 3: Calculate the average speed of mixed traffic flow on the target lane.

[0057] Step 4: Search for time series feature points of average speed of mixed traffic flow on the target lane;

[0058] Step 5: Calculate the total turbulence of mixed traffic flow on the target lane and search for time series feature points of the total turbulence of mixed traffic flow;

[0059] Step 6: Predict the end time T of a single congestion change in the target lane. a ′ and T a Average speed of mixed traffic flow at time ′

[0060] The above steps will be discussed in conjunction with an embodiment of the present invention:

[0061] Step one involves establishing a coordinate system on the urban roads (main roads, secondary roads, and local roads) where congestion prediction is to be performed. The system is defined by taking a specific one-way traffic direction as the direction for which congestion prediction is to be performed. The origin is the intersection of the stop line at the downstream intersection immediately adjacent to the data collection area and the road centerline. The direction for which congestion prediction is to be performed is defined as the positive X-axis, and the positive Y-axis is defined by rotating 90 degrees counterclockwise along the positive X-axis. For example... Figure 2 As shown;

[0062] Data is collected through onboard sensors of intelligent connected vehicles or roadside equipment (road surface induction coils, checkpoint cameras, ultrasonic monitoring devices). and Let t be the time of data acquisition, a be the ID of the intelligent connected vehicle that passes through the data collection area at time t, where a = 1, 2, ..., m(t), and m(t) be the number of intelligent connected vehicles that pass through the data collection area at time t. The instantaneous displacement of the intelligent connected vehicle a at time t consists of two parts, wherein... Let be the x-coordinate of the instantaneous displacement of the intelligent connected vehicle a at time t. Let be the ordinate of the instantaneous displacement of the intelligent connected vehicle a at time t. Let be the instantaneous speed of intelligent connected vehicle a at time t;

[0063] In step one, the data collection area is at least 50 meters long and has a width equal to all Z lanes in the driving direction for which congestion prediction is to be performed. The time interval between the data acquisition time t and the previous time t-1 is no more than 2 seconds. In one embodiment of the present invention, the data collection area is 150 meters long and has a width of 3 lanes in one direction. The road type is a secondary arterial road. The time interval between the data acquisition time t and the previous time t-1 is 1 second. Each subsequent data acquisition time is recorded with 16:30 as the starting point and 18:30 as the ending point (16:30:01 corresponds to the first data acquisition time, 17:00:00 corresponds to the 1800th data acquisition time, 17:30:00 corresponds to 3600s, 18:30:00 corresponds to 5400s, and so on).

[0064] Step two involves calculating the instantaneous displacement and instantaneous speed of non-intelligent connected vehicles. For non-intelligent connected vehicles within the sensor and network sensing range of intelligent connected vehicles, data is collected using onboard sensors or roadside equipment of the intelligent connected vehicles. and Let b be the ID of the non-intelligent connected vehicle that passes through the data collection area at time t, where b = 1, 2, ..., n(t), and n(t) be the number of non-intelligent connected vehicles that pass through the data collection area at time t. The instantaneous displacement of the non-intelligent connected vehicle b at time t consists of two parts, wherein... Let x be the abscissa of the instantaneous displacement of the non-intelligent connected vehicle b at time t. Let be the ordinate of the instantaneous displacement of the non-intelligent connected vehicle b at time t. Let be the instantaneous speed of the non-intelligent connected vehicle b at time t;

[0065] See Figure 3 For non-intelligent connected vehicles that are outside the perception range of intelligent connected vehicles, the distance is estimated based on the instantaneous displacement and instantaneous speed of the nearest intelligent connected vehicles in front and behind. and

[0066]

[0067] in, To and Functions related to q(t) and Q(t), To and Functions related to q(t) and Q(t), for Functions related to q(t), Q(t), q′(t), and Q′(t); at time t, the intelligent connected vehicles adjacent to the non-intelligent connected vehicle numbered b are numbered a, a, and a, respectively. f With a r The non-intelligent connected vehicle numbered b is adjacent to intelligent connected vehicles numbered a on both sides. f ′ and a r At time t, along the X-axis, the intelligent connected vehicle a... f With intelligent connected vehicles a r The number of non-intelligent connected vehicles between them is Q(t), and non-intelligent connected vehicle b is the number of intelligent connected vehicles a. r The nearest non-intelligent connected vehicle q(t); at time t, along the Y-axis, intelligent connected vehicle a f With intelligent connected vehicles a r The number of non-intelligent connected vehicles between points A and B is Q′(t), and non-intelligent connected vehicle b is connected to intelligent connected vehicle a. r The q'(t)th non-intelligent connected vehicle in the nearest neighbor; For intelligent connected vehicles at time t r instantaneous speed, For intelligent connected vehicles at time t r The instantaneous displacement x-coordinate For intelligent connected vehicles at time t r The instantaneous displacement ordinate; For intelligent connected vehicles at time t f instantaneous speed, For intelligent connected vehicles at time t f The instantaneous displacement x-coordinate For intelligent connected vehicles at time t f The instantaneous displacement ordinate, For intelligent connected vehicles at time t r The instantaneous velocity of ′ For intelligent connected vehicles at time t f The instantaneous velocity of ′.

[0068] In this embodiment, further, according to step two... Calculation of q(t), Q(t), q′(t) and Q′(t) and

[0069]

[0070] Then, in step three, the average speed of the mixed traffic flow on the target lane is calculated. Let be the average of the instantaneous speeds of all intelligent connected vehicles and all non-intelligent connected vehicles in the target lane at time t:

[0071]

[0072] A(t) represents the number of mixed traffic vehicles in the target lane at time t, which is the sum of the number of all intelligent connected vehicles m′(t) and the number of all non-intelligent connected vehicles n′(t) in the target lane. A(t) = m′(t) + n′(t). In this embodiment, the target lane is the second lane from the center line of the road to the side, that is, the middle lane among the three lanes in the driving direction for which congestion prediction is to be performed.

[0073] Step four involves searching for time series feature points of the average speed of mixed traffic flow on the target lane: There are a total of four types of time series feature points of the average speed of mixed traffic flow, which are defined as the congestion formation start point T1′(t), the congestion formation end point T2′(t), the congestion dissipation start point T3′(t), and the congestion dissipation end point T4′(t).

[0074] During the search, the time series characteristic points of the average speed of these four types of mixed traffic flows conform to the following mathematical characteristics:

[0075] (1) The starting point of congestion formation, T1′(t), is... The moment when the high-speed fluctuation turns into a monotonically decreasing maximum, where the range of the high-speed fluctuation is... Not less than 40km / h;

[0076] (2) The congestion formation endpoint T2′(t) is: The time corresponding to the minimum point where the monotonically decreasing trend turns into a low-speed fluctuation, where the range of the low-speed fluctuation is... Not exceeding 30km / h;

[0077] (3) The congestion dissipation starting point T3′(t) is The moment when the low-speed fluctuation turns into a monotonically increasing minimum value;

[0078] (4) The congestion dissipation endpoint T4′(t) is The moment when the monotonically increasing trend turns into a high-speed fluctuation at its maximum value;

[0079] The time series of average speed of mixed traffic flow in a target lane includes the number of congestion events and several characteristic points of various types. The complete process of a single congestion event—from the start of congestion to its end, congestion continuation, congestion beginning to dissipate, and congestion completely dissipating—includes one of each of the four types of characteristic points. Furthermore, the order in which the characteristic points of the mixed traffic flow average speed time series appear is T1′(t), T2′(t), T3′(t), and T4′(t). In this embodiment, as shown... Figure 5 As shown;

[0080] Step 5 involves calculating the total turbulence of the mixed traffic flow on the target lane and searching for time series feature points of the total turbulence of the mixed traffic flow.

[0081] See Figure 4 The data collection area is L in length, the number of lanes in the driving direction for which congestion prediction is to be performed is Z, and the target lane is numbered z0. The data collection area is divided into Z x K equal-sized road grids, and the total turbulence E of the mixed traffic flow on the target lane at time t is calculated. S (z0,t):

[0082]

[0083] Where z0 is the number of the road grid along the Y-axis, z0 = 1, 2, ..., Z, k is the number of the road grid along the X-axis, k = 1, 2, ..., K, K is the total number of road grids on a single lane, E C (z0,k,t) represents the mixed traffic flow turbulence on the road grid (z0,k) at time t, E S (z0,t) represents the total turbulence of mixed traffic flow on the target lane at time t, where i is the vehicle number. Let be the distance from the geometric center of vehicle i at time t to the geometric center of the road grid numbered (z0,k). v represents the average vehicle spacing of mixed traffic flow in the data collection area at time t. i (t)cosδ i With v j (t)cosδ j Vehicles i and j at time t are respectively... The component of velocity in direction, v i (t)sinδ i With v j (t)sinδ j Let i and j be perpendicular to each other at time t. Component of velocity in direction, For time t, vehicle i and vehicle j are along The displacement difference in direction, where G represents the vehicle type, G∈{I,N}. At time t, vehicle i is perpendicular to vehicle j. Displacement difference in direction.

[0084] In this embodiment, Z is 3, z0 is 2, and K is 150.

[0085] According to step five, there are four types of time series feature points for total turbulence in mixed traffic flow: the starting point T1(t) of the increase in total turbulence in mixed traffic flow caused by congestion, the ending point T2(t) of the increase in total turbulence in mixed traffic flow caused by congestion, the starting point T3(t) of the increase in total turbulence in mixed traffic flow caused by the dissipation of congestion, and the ending point T4(t) of the increase in total turbulence in mixed traffic flow caused by the dissipation of congestion. During the search, these four types of time series feature points for total turbulence in mixed traffic flow conform to the following mathematical characteristics:

[0086] (1) The starting point of the increase in total turbulence of mixed traffic flow corresponding to congestion is T1(t), which is the congestion formation stage E. S (z0,t) represents the moment corresponding to the turning point where the low-value fluctuations begin to shift towards a significant increase, where the range of the low-value fluctuations is... No more than 300;

[0087] (2) The endpoint T2(t) of the increase in total turbulence of the mixed traffic flow corresponding to the congestion is the congestion formation stage E. S (z0,t) ends its significant increase and turns to the moment corresponding to the turning point of low-value fluctuation;

[0088] (3) The starting point of the increase in total turbulence of mixed traffic flow corresponding to congestion dissipation, T3(t), is the congestion dissipation stage E. S (z0,t) represents the moment when the low-value fluctuations begin to turn into a significant increase.

[0089] (4) The endpoint T4(t) of the total turbulence increase in the mixed traffic flow corresponding to congestion dissipation is the congestion dissipation stage E. S (z0,t) ends its significant increase and turns to the moment corresponding to the turning point of low-value fluctuation.

[0090] A target lane mixed traffic flow total turbulence time series encompasses several feature points of various types. In the complete process of a single congestion event—from the onset of congestion to its end, congestion continuing, congestion beginning to dissipate, and congestion completely dissipating—the corresponding total turbulence time series exhibits a trend of increasing total turbulence, decreasing total turbulence, fluctuating at low values, increasing total turbulence again, and decreasing total turbulence again. This process includes one feature point from each of the four types of total turbulence time series, appearing in the order T1(t), T2(t), T3(t), and T4(t). For the same complete congestion event, the feature points of each type of the target lane mixed traffic flow total turbulence time series always appear before the feature points of the corresponding type in the target lane mixed traffic flow average speed time series. In this embodiment, for example… Figure 5 As shown;

[0091] From step six, predict the end time T of a single congestion change in the target lane. a ′ and T a Average speed of mixed traffic flow at time ′ in:

[0092] E S (z0,t) The time T at which the change begins p Including T1(t) and T3(t); E S (z0,t) The time T at which the change ends a Including T2(t) and T4(t); the starting time T of the target lane congestion change. p Including T1′(t) and T3′(t); the end time T of the congestion change in the target lane. a This includes T2′(t) and T4′(t);

[0093] Average speed of mixed traffic flow at the start of congestion change Including the average speed of mixed traffic flow at the point where congestion begins Average speed of mixed traffic flow corresponding to the congestion dissipation point Average speed of mixed traffic flow at the end of congestion change Including the average speed of mixed traffic flow at the endpoint of congestion formation Average speed of mixed traffic flow corresponding to the congestion dissipation endpoint

[0094] By observing T p T p ′、T a , And calculate E at each time point S Solve for (z,t) T2′(t) and T4′(t):

[0095]

[0096] Where c w With c w ′ represents the polynomial coefficients, w = 0, 1, 2, 3, 4; ΔE S (z0,t) represents E at time t. S The change in (z0,t) compared to the previous moment, ΔE S (z0,t)=E S (z0,t)-E S (z0,t-1); s is the average velocity of the mixed traffic flow in the monotonic interval [T]. p ′,T a The slope on ′].

[0097] Furthermore, in step six, s is approximately equal to the average speed of the mixed traffic flow in the monotonic interval [T]. p ′,T a The slope on ′], i.e.

[0098] In this embodiment, the coefficients of the polynomial characterizing the relationship between the average speed time series of a single-lane mixed traffic flow and the total turbulence time series of a single-lane mixed traffic flow are calibrated using the least squares method, and T is recorded. p T p ′、T a , Observations (unit: m / s) and summaries c w With c w The calibration results of ′ are shown in Table 1; and The observed values ​​are 6.79 m / s and 10.41 m / s, respectively. Based on step six, the approximate value of s in this embodiment is 7.67 × 10⁻⁶. -3 With 6.38×10 -3 Substitute ΔE S (z0, t), obtained according to the above steps in this embodiment The prediction results of T2′(t) and T4′(t) are shown in Table 2.

[0099] Table 1. Observed values ​​of each parameter and calibration results of polynomial coefficients

[0100]

[0101] Table 2. Congestion change endpoints and corresponding average speed predictions for mixed traffic flows.

[0102]

[0103] In summary, this invention proposes a lane-level congestion prediction method for urban intelligent connected mixed traffic flow. In practical applications, based on the level of urban road intelligence and the penetration rate of intelligent connected vehicles in mixed traffic flow, and supplemented by roadside equipment such as road induction coils, checkpoint cameras, and ultrasonic monitoring devices to collect real-time vehicle trajectory information, the method predicts the endpoint of congestion formation, the endpoint of congestion dissipation, and the corresponding average speed of the mixed traffic flow. This provides a theoretical basis and model support for lane-level congestion prediction and traffic flow organization for urban intelligent connected mixed traffic flow, enabling timely traffic control measures to prevent and alleviate congestion, thereby improving the operational efficiency and safety of the transportation system.

Claims

1. A lane-level congestion prediction method for urban intelligent connected mixed traffic flow, characterized in that, Includes the following steps: Step 1: Obtain the instantaneous displacement and instantaneous speed of the intelligent connected vehicle; A coordinate system is established with the intersection of the stop line and the center line of the road at the downstream intersection adjacent to the data collection area as the origin, the driving direction for which congestion prediction is to be performed as the positive X-axis, and the positive Y-axis as the direction of rotation 90 degrees counterclockwise along the positive X-axis. Data collected via vehicle-mounted or roadside equipment and Let t be the time of data acquisition, a be the ID of the intelligent connected vehicle that passes through the data collection area at time t, where a = 1, 2, ..., m(t), and m(t) be the number of intelligent connected vehicles that pass through the data collection area at time t. The instantaneous displacement of the intelligent connected vehicle a at time t consists of two parts, wherein... Let be the x-coordinate of the instantaneous displacement of the intelligent connected vehicle a at time t. Let be the ordinate of the instantaneous displacement of the intelligent connected vehicle a at time t. Let be the instantaneous speed of intelligent connected vehicle a at time t; Step 2: Calculate the instantaneous displacement and instantaneous speed of the non-intelligent connected vehicle. Let b be the ID of the non-intelligent connected vehicle that passes through the data collection area at time t, where b = 1, 2, ..., n(t), and n(t) be the number of non-intelligent connected vehicles that pass through the data collection area at time t. The instantaneous displacement of the non-intelligent connected vehicle b at time t consists of two parts, wherein... Let x be the abscissa of the instantaneous displacement of the non-intelligent connected vehicle b at time t. Let be the ordinate of the instantaneous displacement of the non-intelligent connected vehicle b at time t. Let be the instantaneous speed of the non-intelligent connected vehicle b at time t; For non-intelligent connected vehicles within the perception range of intelligent connected vehicles, data is collected through onboard or roadside equipment. and For non-intelligent connected vehicles that are outside the perception range of intelligent connected vehicles, the distance is estimated based on the instantaneous displacement and instantaneous speed of the nearest intelligent connected vehicles in front and behind. and in, To and Functions related to q(t) and Q(t), To and Functions related to q(t) and Q(t), for Functions related to q(t), Q(t), q′(t), and Q′(t); at time t, the intelligent connected vehicles adjacent to the non-intelligent connected vehicle numbered b are numbered a, a, and a, respectively. f With a r The non-intelligent connected vehicle numbered b is adjacent to intelligent connected vehicles numbered a on both sides. f ′ and a r At time t, along the X-axis, the intelligent connected vehicle a... f With intelligent connected vehicles a r The number of non-intelligent connected vehicles between them is Q(t), and non-intelligent connected vehicle b is the number of intelligent connected vehicles a. r The nearest non-intelligent connected vehicle q(t); at time t, along the Y-axis, intelligent connected vehicle a f With intelligent connected vehicles a r The number of non-intelligent connected vehicles between points A and B is Q′(t), and non-intelligent connected vehicle b is connected to intelligent connected vehicle a. r The q'(t)th non-intelligent connected vehicle in the nearest neighbor; For intelligent connected vehicles at time t r instantaneous speed, For intelligent connected vehicles at time t r The instantaneous displacement x-coordinate For intelligent connected vehicles at time t r The instantaneous displacement ordinate; For intelligent connected vehicles at time t f instantaneous speed, For intelligent connected vehicles at time t f The instantaneous displacement x-coordinate For intelligent connected vehicles at time t f The instantaneous displacement ordinate, For intelligent connected vehicles at time t r The instantaneous velocity of ′ For intelligent connected vehicles at time t f The instantaneous velocity of ′; Step 3: Calculate the average speed of mixed traffic flow on the target lane. Let be the average of the instantaneous speeds of all intelligent connected vehicles and all non-intelligent connected vehicles in the target lane at time t: A(t) is the number of mixed traffic vehicles in the target lane at time t, which is the sum of the number of all intelligent connected vehicles m′(t) and the number of all non-intelligent connected vehicles n′(t) in the target lane. A(t) = m′(t) + n′(t). Step 4: Search for time series feature points of average speed of mixed traffic flow on the target lane; There are four categories of time series characteristic points for the average speed of mixed traffic flow, defined as the congestion initiation point T′1(t), the congestion initiation point T2′(t), the congestion dissipation initiation point T3′(t), and the congestion dissipation point T4′(t). During the search, these four categories of time series characteristic points for the average speed of mixed traffic flow conform to the following mathematical characteristics: (1) The starting point of congestion formation is T1′(t) The moment when the high-speed fluctuation turns into a monotonically decreasing maximum, where the range of the high-speed fluctuation is... Not less than 40km / h; (2) The congestion formation endpoint T2′(t) is: The time corresponding to the minimum point where the monotonically decreasing trend turns into a low-speed fluctuation, where the range of the low-speed fluctuation is... Not exceeding 30km / h; (3) The congestion dissipation starting point T3′(t) is The moment when the low-speed fluctuation turns into a monotonically increasing minimum value; (4) The congestion dissipation endpoint T4′(t) is The moment when the monotonically increasing trend turns into a high-speed fluctuation at its maximum value; Step 5: Calculate the total turbulence of mixed traffic flow on the target lane and search for time series feature points of the total turbulence of mixed traffic flow; The data collection area is L in length, the number of lanes in the driving direction for which congestion prediction is to be performed is Z, and the target lane is numbered z0. The data collection area is divided into Z x K equal-sized road grids, and the total turbulence E of the mixed traffic flow on the target lane at time t is calculated. S (z0,t): Where z0 is the number of the road grid along the Y-axis, z0 = 1, 2, ..., Z, k is the number of the road grid along the X-axis, k = 1, 2, ..., K, K is the total number of road grids on a single lane, E C (z0,k,t) represents the mixed traffic flow turbulence on the road grid (z0,k) at time t, E S (z0,t) represents the total turbulence of mixed traffic flow on the target lane at time t, where i is the vehicle number and d is the total turbulence. z0,k,i (t) represents the distance from the geometric center of vehicle i to the geometric center of the road grid numbered (z0,k) at time t. v represents the average vehicle spacing of mixed traffic flow in the data collection area at time t. i (t)cosδ i With v j (t)cosδ j Vehicles i and j at time t are respectively... The component of velocity in direction, v i (t)sinδ i With v j (t)sinδ j Let i and j be perpendicular to each other at time t. Component of velocity in direction, For time t, vehicle i and vehicle j are along The displacement difference in direction, where G represents the vehicle type, G∈{I,N}. At time t, vehicle i is perpendicular to vehicle j. Displacement difference in direction; There are four categories of time series feature points for total turbulence in mixed traffic flow: T1(t) corresponding to the starting point of the increase in total turbulence caused by congestion, T2(t) corresponding to the ending point of the increase in total turbulence caused by congestion, T3(t) corresponding to the starting point of the increase in total turbulence caused by the dissipation of congestion, and T4(t) corresponding to the ending point of the increase in total turbulence caused by the dissipation of congestion. During the search, these four categories of time series feature points for total turbulence in mixed traffic flow conform to the following mathematical characteristics: (1) The starting point of the increase in total turbulence of mixed traffic flow corresponding to congestion is T1(t), which is the congestion formation stage E. S (z0,t) represents the moment corresponding to the turning point where the low-value fluctuations begin to shift towards a significant increase, where the range of the low-value fluctuations is... No more than 300; (2) The endpoint T2(t) of the increase in total turbulence of the mixed traffic flow corresponding to the congestion is the congestion formation stage E. S (z0,t) ends its significant increase and turns to the moment corresponding to the turning point of low-value fluctuation; (3) The starting point of the increase in total turbulence of mixed traffic flow corresponding to congestion dissipation, T3(t), is the congestion dissipation stage E. S (z0,t) represents the moment when the low-value fluctuations begin to turn into a significant increase. (4) The endpoint T4(t) of the total turbulence increase in the mixed traffic flow corresponding to congestion dissipation is the congestion dissipation stage E. S (z0,t) ends its significant increase and turns to the moment corresponding to the turning point of low-value fluctuation; Step 6: Predict the end time T of a single congestion change in the target lane. a ′ and T a Average speed of mixed traffic flow at time ′ E S (z0,t) The time T at which the change begins p Including T1(t) and T3(t); E S (z0,t) The time T at which the change ends a Including T2(t) and T4(t); the starting time T of the target lane congestion change. p Including T1′(t) and T3′(t); the end time T of the congestion change in the target lane. a This includes T2′(t) and T4′(t); Average speed of mixed traffic flow at the start of congestion change Including the average speed of mixed traffic flow at the point where congestion begins Average speed of mixed traffic flow corresponding to the congestion dissipation point Average speed of mixed traffic flow at the end of congestion change Including the average speed of mixed traffic flow at the endpoint of congestion formation Average speed of mixed traffic flow corresponding to the congestion dissipation endpoint By observing T p T p ′、T a , And calculate E at each time point S Solve for (z,t) T2′(t) and T4′(t): Where c w With c w ′ represents the polynomial coefficients, w = 0, 1, 2, 3, 4; ΔE S (z0,t) represents E at time t. S The change in (z0,t) compared to the previous moment, ΔE S (z0,t)=E S (z0,t)-E S (z0,t-1); s is the average velocity of the mixed traffic flow in the monotonic interval [T]. p ′,T a The slope on ′].

2. The lane-level congestion prediction method for urban intelligent connected mixed traffic flow according to claim 1, characterized in that, The data collection area is no less than 50 meters long and the width is the width of all Z lanes in the driving direction for which congestion prediction is to be performed. The time interval between the data acquisition time t and the previous time t-1 is no more than 2 seconds.

3. The lane-level congestion prediction method for urban intelligent connected mixed traffic flow according to claim 1, characterized in that, according to Calculation of q(t), Q(t), q′(t) and Q′(t) and The specific method is as follows:

4. A lane-level congestion prediction method for urban intelligent connected mixed traffic flow according to claim 1, characterized in that, In the calculation, s is approximately equal to the average speed of the mixed traffic flow in the monotonic interval [T]. p ′,T a The slope on ], i.e.

Images