Method for identifying traffic conflicts in roundabout based on curve driving behaviors

Abstract

The invention discloses a method for identifying traffic conflicts in a roundabout based on curve driving behavior characteristics, which relates to the technical field of intelligent traffic management and comprises the following steps: acquiring channelization data and vehicle video information of the roundabout; accurately positioning the positions of the target vehicle and the adjacent vehicle to obtain motion information; each future movement track is predicted, the TTC of the target vehicle and the adjacent vehicle at any moment is calculated, and when the TTC value is lower than a set safety threshold value, the system judges that a conflict risk exists; vehicle conflict analysis and early warning at the annular intersection can be realized, a driver is reminded in time, and rear-end collision is avoided.

Description

Technical Field

[0001] The invention relates to the technical field of intelligent traffic management, and in particular to a method for identifying traffic conflicts in a roundabout based on curve driving behavior. Background Art

[0002] With the acceleration of urbanization, traffic problems are becoming increasingly prominent, especially at intersections, which are key nodes in the urban road traffic network. They bear the functions of gathering vehicles and pedestrians and changing directions. The instantaneous traffic is concentrated within the intersection, there are many intersection conflict points, and the driving tasks are complex. It is a bottleneck of road traffic capacity and an accident-prone area. How to achieve early warning of vehicle conflicts at intersections is an important topic.

[0003] Currently, most of the research on intersection vehicle conflict warning is aimed at crossroads, but there is little research on roundabouts. The vehicle movement rules at crossroads and roundabouts are different, and traffic conflicts are also different from those at crossroads.

[0004] Based on this, the present invention aims to provide a method for early warning of conflicts at roundabouts, which can provide real-time early warning of traffic conflicts at roundabouts. Summary of the invention

[0005] In order to make up for the shortcomings of the existing technical problems, the purpose of the present invention is to provide a traffic conflict identification method in a roundabout based on curved driving behavior, obtain vehicle video information of the roundabout, analyze the vehicle data in the video, predict the future movement trajectory of the vehicle, perform conflict analysis and early warning on future vehicles, and promptly remind the driver to avoid rear-end collisions.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for identifying traffic conflicts in a roundabout based on curved driving behavior comprises the following steps:

[0008] (1) Obtain intersection channelization data, use a high-altitude camera to take panoramic photos of the intersection, and collect video information of vehicles at the intersection;

[0009] (2) Analyze the video in step (1) using target detection and multi-target tracking technology, and analyze the continuous frames of the video to accurately locate the position of the target vehicle and adjacent vehicles and obtain motion information;

[0010] (3) According to step (2), the position information and motion information of the target vehicle and the adjacent vehicles are obtained to predict their respective future motion trajectories, and the TTC of the target vehicle and the adjacent vehicles is calculated at any time. When the TTC value is lower than the set safety threshold, the system will determine that there is a risk of collision.

[0011] In the present invention, the current positions of the target vehicle and its front and rear vehicles, as well as the motion information, are located in step (2), specifically as follows:

[0012] A coordinate system is established with the coordinate center of the center of the roundabout to determine the coordinates of each exit;

[0013] The speed, acceleration, driving angle and angular velocity of the target vehicle and its adjacent vehicles are calculated based on their current positions and the positions of the previous frame, as follows:

[0014] The positions of the target vehicle at the current moment and the previous frame are (x n1 ,y n1 )、(x n0 ,y n0 ); the positions of the adjacent vehicles at the current moment and in the previous frame are respectively (x m1 ,y m1 )、(x m0 ,y m0 );

[0015] The speed of the target vehicle along the x-axis is:

[0016] Among them, Δt is the time difference between the current moment and the previous frame;

[0017] The speed of the target vehicle along the y-axis is:

[0018] The current speed of the target vehicle is:

[0019] The current driving angle of the target vehicle: θ n1 =atan2(y n1 -y n0 ,x n1 -x n0 ), where atan2 is the arc tangent angle from the X axis to the position;

[0020] The angular velocity of the target vehicle at the current moment:

[0021] The speed of the adjacent vehicle along the x-axis is: Among them, Δt is the time difference between the current moment and the previous frame;

[0022] The velocity of the adjacent vehicle along the y-axis is:

[0023] The current speed of the adjacent vehicle is:

[0024] The driving angle of the adjacent vehicle at the current moment: θ m1 =atan2(y m1 -y m0 ,x m1 -x m0 ), where atan2 is the arc tangent angle from the X axis to the position;

[0025] The angular velocity of the adjacent vehicle at the current moment:

[0026] In the present invention, the future motion trajectories of the target vehicle and the adjacent vehicles are as follows:

[0027] Determine the position of the target vehicle and other vehicles at time t

[0028] The position of the target vehicle n at time t in the future is:

[0029]

[0030] x n1 ,y n1 are the coordinates of the target vehicle at the current moment, θ n1 is the driving angle of the target vehicle at the current moment; ω n1 is the angular velocity of the target vehicle at the current moment;

[0031] Calculate the position x of the target vehicle n at the next time t-1 n (t-1),y n (t-1), calculate the speed v of the target vehicle n at time t n (t);

[0032]

[0033] The position of the adjacent vehicle m at time t is:

[0034]

[0035] x m1 ,y m1 are the coordinates of the adjacent vehicles at the current moment, θ m1 is the driving angle of the adjacent vehicle at the current moment; ω m1 is the angular velocity of the adjacent vehicle at the current moment;

[0036] When the target vehicle n is in front of the adjacent vehicle m, n (t), y n (t) Constraints are applied to ensure that the target vehicle n is located within the roundabout:

[0037] Among them, x e,y e is the closest exit coordinate in front of the target vehicle n;

[0038] The adjacent vehicle m is located in front of the target vehicle n. m (t), y m (t) Constraints are applied to ensure that the adjacent vehicle m is located within the roundabout:

[0039] Among them, x e ,y e is the closest exit coordinate in front of the adjacent vehicle m;

[0040] Calculate the position x of the adjacent vehicle m at the next t-1 time m (t-1),y m (t-1), calculate the velocity v of the adjacent vehicle m at time t m (t);

[0041]

[0042] In the present invention, the steps for calculating the TTC of the target vehicle and the adjacent vehicles at any time are as follows:

[0043]

[0044] Among them, R rel is the average radius of the roundabout; then θ n (t),θ m (t) are the driving angles of the target vehicle n and the adjacent vehicle m at time t; v n (t), v m (t) are the angular velocities of the target vehicle n and the adjacent vehicle m at time t;

[0045]

[0046] The radius of the target vehicle n at time t:

[0047] The radius of the adjacent vehicle m at time t:

[0048]

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The collision warning method of the present invention is targeted at roundabouts, which can predict the future trajectory of the vehicle according to the curved driving behavior of the vehicle at the roundabout, and predict whether the vehicle will have a collision at a future moment.

[0051] 2. The present invention utilizes a high-altitude camera to obtain real-time video information of vehicles at intersections, analyzes the video information of vehicles at intersections, and uses the TTC of the predicted future motion trajectory of vehicles to issue conflict warnings, thereby gaining more driver reaction time and avoiding traffic accidents caused by untimely reactions.

[0052] 3. The present invention utilizes a high-altitude camera to obtain real-time video information of vehicles at intersections and to perform dynamic conflict prediction of TTC in seconds. The dynamic real-time feedback in seconds can not only provide data for predicting the motion trajectory of vehicles, but also respond to sudden changes in vehicle motion, timely correct predictions, and improve prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the roundabout structure of the present invention. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0055] like Figure 1 As shown, the present invention discloses a method for identifying traffic conflicts in a roundabout based on curved driving behavior, comprising the following steps:

[0056] A method for identifying traffic conflicts in a roundabout based on curved driving behavior comprises the following steps:

[0057] (1) Obtain intersection channelization data, use a high-altitude camera to take panoramic photos of the intersection, and collect video information of vehicles at the intersection;

[0058] (2) The video data information collected in step (1) will be transmitted to the background in a timely manner, and the video in step (1) will be analyzed using target detection and multi-target tracking technology, and the target vehicle and adjacent vehicles will be accurately located through continuous frame analysis to obtain motion information;

[0059] Locate the current position of the target vehicle and the vehicles before and after it, as well as its motion information, as follows:

[0060] A coordinate system is established with the coordinate center of the center of the roundabout to determine the coordinates of each exit;

[0061] The speed, acceleration, driving angle and angular velocity of the target vehicle and its adjacent vehicles are calculated based on their current positions and the positions of the previous frame, as follows:

[0062] The positions of the target vehicle at the current moment and the previous frame are (xn1 ,y n1 )、(x n0 ,y n0 ); the positions of the adjacent vehicles at the current moment and in the previous frame are respectively (x m1 ,y m1 )、(x m0 ,y m0 );

[0063] The speed of the target vehicle along the x-axis is:

[0064] Among them, Δt is the time difference between the current moment and the previous frame;

[0065] The speed of the target vehicle along the y-axis is:

[0066] The current speed of the target vehicle is:

[0067] The current driving angle of the target vehicle: θ n1 =atan2(y n1 -y n0 ,x n1 -x n0 ), where atan2 is the arc tangent angle from the X axis to the position;

[0068] The angular velocity of the target vehicle at the current moment:

[0069] The speed of the adjacent vehicle along the x-axis is: Among them, Δt is the time difference between the current moment and the previous frame;

[0070] The velocity of the adjacent vehicle along the y-axis is:

[0071] The current speed of the adjacent vehicle is:

[0072] The driving angle of the adjacent vehicle at the current moment: θ m1 =atan2(y m1 -y m0 ,x m1 -x m0 ), where atan2 is the arc tangent angle from the X axis to the position;

[0073] The angular velocity of the adjacent vehicle at the current moment:

[0074] (3) According to step (2), the position information and motion information of the target vehicle and the adjacent vehicles are obtained to predict their respective future motion trajectories, and the TTC of the target vehicle and the adjacent vehicles is calculated at any time. When the TTC value is lower than the set safety threshold, the system will determine that there is a risk of collision.

[0075] The future motion trajectories of the target vehicle and the adjacent vehicles are as follows:

[0076] Determine the position of the target vehicle and other vehicles at time t

[0077] The position of the target vehicle n at time t in the future is:

[0078]

[0079] x n1 ,y n1 are the coordinates of the target vehicle at the current moment, θ n1 is the driving angle of the target vehicle at the current moment; ω n1 is the angular velocity of the target vehicle at the current moment;

[0080] Calculate the position x of the target vehicle n at the next time t-1 n (t-1),y n (t-1), calculate the speed v of the target vehicle n at time t n (t);

[0081]

[0082] The position of the adjacent vehicle m at time t is:

[0083]

[0084] x m1 ,y m1 are the coordinates of the adjacent vehicles at the current moment, θ m1 is the driving angle of the adjacent vehicle at the current moment; ω m1 is the angular velocity of the adjacent vehicle at the current moment;

[0085] When the target vehicle n is in front of the adjacent vehicle m, n (t), y n (t) Constraints are applied to ensure that the target vehicle n is located within the roundabout:

[0086] Among them, x e ,y e is the closest exit coordinate in front of the target vehicle n;

[0087] The adjacent vehicle m is located in front of the target vehicle n. m(t), y m (t) Constraints are applied to ensure that the adjacent vehicle m is located within the roundabout:

[0088] Among them, x e ,y e is the closest exit coordinate in front of the adjacent vehicle m;

[0089] By constraining the target vehicle n or the adjacent vehicle m, the vehicle can be ensured to be in the roundabout. If the vehicle does not leave at the nearest exit, the next frame of video information will continue the conflict analysis.

[0090] Calculate the position x of the adjacent vehicle m at the next t-1 time m (t-1),y m (t-1), calculate the velocity v of the adjacent vehicle m at time t m (t);

[0091]

[0092] In the present invention, the steps for calculating the TTC of the target vehicle and the adjacent vehicles at any time are as follows:

[0093]

[0094] Among them, R rel is the average radius of the roundabout; then θ n (t),θ m (t) are the driving angles of the target vehicle n and the adjacent vehicle m at time t; v n (t), v m (t) are the angular velocities of the target vehicle n and the adjacent vehicle m at time t;

[0095]

[0096] The radius of the target vehicle n at time t:

[0097] The radius of the adjacent vehicle m at time t:

[0098]

[0099] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for identifying traffic conflicts in roundabouts based on curved driving behavior, characterized in that: The following steps are involved: (1) Obtain intersection channelization data, use a high-altitude camera to take panoramic photos of the intersection, and collect video information of vehicles at the intersection; (2) Analyze the video in step (1) using target detection and multi-target tracking technology, and analyze the continuous frames of the video to accurately locate the position of the target vehicle and adjacent vehicles and obtain motion information; (3) According to step (2), the position information and motion information of the target vehicle and the adjacent vehicles are obtained to predict their respective future motion trajectories, and the TTC of the target vehicle and the adjacent vehicles is calculated at any time. When the TTC value is lower than the set safety threshold, the system will determine that there is a risk of collision.

2. The method for identifying traffic conflicts in roundabouts based on curved driving behavior according to claim 1, characterized in that: In step (2), the current positions of the target vehicle and the vehicles before and after it, as well as the motion information, are located as follows: A coordinate system is established with the coordinate center of the center of the roundabout to determine the coordinates of each exit; The speed, acceleration, driving angle and angular velocity of the target vehicle and its adjacent vehicles are calculated based on their current positions and the positions of the previous frame, as follows: The positions of the target vehicle at the current moment and the previous frame are (x n1 ,y n1 )、(x n0 ,y n0 ); the positions of the adjacent vehicles at the current moment and in the previous frame are respectively (x m1 ,y m1 )、(x m0 ,y m0 ); The speed of the target vehicle along the x-axis is: Among them, Δt is the time difference between the current moment and the previous frame; The speed of the target vehicle along the y-axis is: The current speed of the target vehicle is: The current driving angle of the target vehicle: θ n1 =atan2(y n1 -y n0 ,x n1 -x n0 ), where atan2 is the arc tangent angle from the X axis to the position; The angular velocity of the target vehicle at the current moment: The speed of the adjacent vehicle along the x-axis is: Among them, Δt is the time difference between the current moment and the previous frame; The velocity of the adjacent vehicle along the y-axis is: The current speed of the adjacent vehicle is: The driving angle of the adjacent vehicle at the current moment: θ m1 =atan2(y m1 -y m0 ,x m1 -x m0 ), where atan2 is the arc tangent angle from the X axis to the position; The angular velocity of the adjacent vehicle at the current moment:

3. The method for identifying traffic conflicts in roundabouts based on curved driving behavior according to claim 1, characterized in that: The future motion trajectories of the target vehicle and the adjacent vehicles are as follows: Determine the position of the target vehicle and other vehicles at time t The position of the target vehicle n at time t in the future is: x n1 ,y n1 are the coordinates of the target vehicle at the current moment, θ n1 is the driving angle of the target vehicle at the current moment; ω n1 is the angular velocity of the target vehicle at the current moment; Calculate the position x of the target vehicle n at the next time t-1 n (t-1),y n (t-1), calculate the speed v of the target vehicle n at time t n (t); The position of the adjacent vehicle m at time t is: x m1 ,y m1 are the coordinates of the adjacent vehicle m at the current moment, θ m1 is the driving angle of the adjacent vehicle at the current moment; ω m1 is the angular velocity of the adjacent vehicle at the current moment; When the target vehicle n is in front of the adjacent vehicle m, n (t), y n (t) Constraints are applied to ensure that the target vehicle n is located within the roundabout: Among them, x e ,y e is the closest exit coordinate in front of the target vehicle n; The adjacent vehicle m is located in front of the target vehicle n. m (t), y m (t) Constraints are applied to ensure that the adjacent vehicle m is located within the roundabout: Among them, x e ,y e is the closest exit coordinate in front of the adjacent vehicle m; Calculate the position x of the adjacent vehicle m at the next t-1 time m (t-1),y m (t-1), calculate the velocity v of the adjacent vehicle m at time t m (t); 4. The method for identifying traffic conflicts in roundabouts based on curved driving behavior according to claim 1, characterized in that: The calculation steps of the TTC between the target vehicle and the adjacent vehicles at any time are as follows: Among them, R rel is the average radius of the roundabout; then θ n (t),θ m (t) are the driving angles of the target vehicle n and the adjacent vehicle m at time t; v n (t), v m (t) are the angular velocities of the target vehicle n and the adjacent vehicle m at time t; The radius of the target vehicle n at time t: The radius of the adjacent vehicle m at time t:

Images