Signal intersection timing parameter estimation method and system based on UWB high-precision positioning data

Abstract

The invention provides a signal intersection timing parameter estimation method and system based on UWB high-precision positioning data, and the method comprises the steps: obtaining the high-precision space-time trajectory data of a plurality of passing vehicles through a UWB base station network disposed in an intersection region, judging the vehicle passing mode based on the time-speed change curve of the passing vehicles, and carrying out the estimation of the timing parameters of the signal intersection. Respectively extracting a parking starting moment and a passing moment of each passing vehicle; constructing a passing event sequence containing the parking starting time and the passing time of all passing vehicles; preliminarily dividing a signal period based on the traffic event sequence; and constructing a signal timing model defined by the preliminarily divided signal period, the green light initial offset and the effective green light duration, and performing fine adjustment on timing parameters in the signal timing model by constructing an optimization objective function to obtain an optimal timing parameter solution. By means of the method and the corresponding system, timing misjudgment caused by abnormal driving and data noise is effectively restrained, and timing estimation accuracy is improved.

Description

Technical Field

[0001] This invention proposes a method and system for estimating timing parameters of signalized intersections based on UWB high-precision positioning data, which relates to the field of urban traffic data analysis technology. Background Technology

[0002] In urban traffic management, the timing scheme of signalized intersections (including cycle length, green light start and end time, phase sequence, etc.) is a key parameter for optimizing traffic efficiency and reducing delays and emissions. Traditional methods of obtaining timing information rely on the communication interface of the signal controller (such as the NTCIP protocol) or manual surveys, which have the following problems: (1) Old signal controllers lack data interfaces and cannot remotely obtain real-time timing; (2) Manual observation is costly and inaccurate, and it is difficult to cover multiple time periods and multiple intersections; (3) Existing methods based on GPS or floating vehicle data are limited by positioning accuracy and cannot accurately identify the stopping and starting behavior of vehicles before the stop line, resulting in large timing estimation errors; (4) Video or geomagnetic detectors are costly to deploy and are easily affected by weather and obstruction. Most existing timing inversion algorithms based on floating car data assume that the behavior of vehicles passing through intersections is idealized or homogeneous, and lack in-depth mining of vehicle micro-dynamic characteristics (such as instantaneous speed and acceleration change rate); (5) In actual traffic scenarios, there are common non-standard traffic behaviors such as drivers being distracted and slow to start from the green light, aggressive driving to rush through the yellow light, or hesitating and slowing down during the green light due to unclear road conditions. If the technology cannot effectively identify and remove these abnormal samples containing high noise, it will often lead to significant deviations in the deduced green light start and end times, reducing the robustness of timing parameter estimation.

[0003] In recent years, UWB technology has been widely used in indoor positioning and vehicle-road cooperative scenarios due to its centimeter-level positioning accuracy, high sampling rate, and strong anti-interference capability. However, there is currently no systematic method for using UWB trajectory data for signal timing inversion. Therefore, there is an urgent need for a technical solution that can estimate intersection timing parameters with high accuracy based solely on UWB vehicle trajectory data, without relying on signal communication. Summary of the Invention

[0004] This invention provides a method and system for estimating timing parameters of signalized intersections based on UWB high-precision positioning data, in order to solve the problems mentioned above: The present invention proposes a method for estimating signal intersection timing parameters based on UWB high-precision positioning data, the method comprising: By deploying a UWB base station network in the intersection area, high-precision spatiotemporal trajectory data of multiple passing vehicles is obtained. The high-precision spatiotemporal trajectory data includes timestamp t, two-dimensional coordinates (x, y) and instantaneous speed v. Based on the high-precision spatiotemporal trajectory data and the preset geographic coordinates of the stop lines of each entrance lane, the time and time-speed change curve of each passing vehicle crossing the stop line are obtained. Based on the time-speed change curve, the vehicle traffic pattern is determined, and based on the vehicle traffic pattern, the stopping and starting time and the passing time of each passing vehicle are extracted respectively; The stopping and starting times and passing times of all passing vehicles are merged to construct a passing event sequence that includes both stopping and starting times and passing times; The passage event sequence is subjected to time interval analysis to preliminarily divide the signal period; Under the initially defined signal period, the time in the passage event sequence is mapped to the interval [0, C), where C represents the initially defined signal period; Clustering algorithms are used to cluster the time sequences of traffic events mapped to the signal period, identifying densely populated traffic event intervals as the green light phase window. , ),in, Indicates the initial offset of the green light. Indicates the effective green light duration; A signal timing model is constructed based on the initially defined signal period, green light start offset, and effective green light duration. The timing parameters in the signal timing model are then fine-tuned by constructing an optimization objective function to obtain the optimal timing parameter solution.

[0005] Furthermore, based on the high-precision spatiotemporal trajectory data and the preset geographical coordinates of the stop lines at each approach lane, the time and time-speed variation curve of each vehicle crossing the stop line are obtained, including: Preset the geographical coordinates of the stop lines at each entrance lane; Using timestamp t as the independent variable, cubic spline interpolation is performed on the discrete two-dimensional coordinates (x,y) of each passing vehicle to construct parametric equations x(t) and y(t) of the position of each passing vehicle with respect to time. The first derivatives of the parametric equations are taken to obtain the component velocities, and the continuous time-velocity variation curves of each passing vehicle are synthesized. And based on the parametric equation, the moment when the coordinates of each passing vehicle coincide with the coordinates of the stop line is obtained.

[0006] Furthermore, based on the time-speed variation curve, the vehicle traffic pattern is determined, and based on the vehicle traffic pattern, the stopping and starting time and passing time of each passing vehicle are extracted, including: Based on the time-speed change curve, determine whether the speed of the passing vehicle drops below a preset speed threshold before the stop line and remains above a preset time threshold. If so, mark it as a red light stop, record the time from its stationary state to its acceleration, and define that time as the stop-start time of the passing vehicle. Otherwise, mark the vehicle as a green light through vehicle and record the time it crosses the stop line.

[0007] Furthermore, the clustering algorithms include the DBSCAN clustering algorithm and the K-means clustering algorithm.

[0008] Furthermore, the optimization objective function is constructed based on the conflict loss function, which is obtained according to the following rules: A high-weight loss occurs when a vehicle passes through a stop line without stopping and the time of its crossing is within the red light period of the signal timing model, or when a vehicle stops but the time of its stop-start is within the red light period of the signal timing model; otherwise, a low-weight loss occurs.

[0009] Furthermore, the specific form of the conflict loss function is as follows: in, This represents the function value of the conflict loss function. Indicates the first The loss caused by a green light through bus. Indicates the first The loss caused by a vehicle stopping at a red light. Indicates the weighting coefficient. Indicates the first The weight of a green light express lane Indicates the first The weight of a vehicle stopping at a red light. This indicates the group of vehicles that have been designated as "green light" through traffic. This indicates the group of vehicles that have been deemed to be stopped at a red light.

[0010] Furthermore, in, A function indicating the green light for through traffic; = in, Function to indicate a green light when stopping at a red light in, Indicates the first The time when a green-light through vehicle crosses the stop line; in, Indicates the first The moment a vehicle stops and starts moving when stopped at a red light.

[0011] Furthermore, the green light phase window consists of the following three consecutive time regions: Green light start blur zone Clearly define the green zone The blurry area ends with the green light The green light initial ambiguity zone is designed to tolerate vehicle start-up delays; vehicles should be within the clearly defined green light zone. Normal passage, the green light ends the ambiguous area. This is to allow vehicles to pass through at the end of the green light period.

[0012] Furthermore, the optimization objective function is: in, This represents the optimal timing parameter estimation solution.

[0013] The present invention proposes a signal intersection timing parameter estimation system based on UWB high-precision positioning data, the system comprising: The spatiotemporal trajectory data acquisition module is used to acquire high-precision spatiotemporal trajectory data of multiple passing vehicles through a UWB base station network deployed in the intersection area. The high-precision spatiotemporal trajectory data includes timestamp t, two-dimensional coordinates (x, y) and instantaneous speed v. The module for acquiring time-velocity change curves acquires the time and time-velocity change curve of each vehicle crossing the stop line based on the high-precision spatiotemporal trajectory data and the preset geographic coordinates of the stop lines of each entrance lane. The vehicle traffic mode determination module determines the vehicle traffic mode based on the time-speed change curve, and extracts the stopping and starting time and passing time of each passing vehicle based on the vehicle traffic mode; A passage event sequence module is constructed to merge the stop start time and passage time of all passing vehicles and construct a passage event sequence containing the stop start time and passage time; A preliminary signal period segmentation module is used to perform time interval analysis on the passage event sequence in order to initially segment the signal period. The time mapping module is used to map the time in the passage event sequence to the interval [0, C) under the initially divided signal period; The clustering module is used to cluster the times in the sequence of traffic events mapped to the signal period using a clustering algorithm, identify dense intervals of traffic events, and use them as the green light phase window. , ),in, Indicates the initial offset of the green light. Indicates the effective green light duration; The optimal solution module is used to construct a signal timing model defined by the initially divided signal period, green light start offset, and effective green light duration, and to fine-tune the timing parameters in the signal timing model by constructing an optimization objective function to obtain the optimal timing parameter solution.

[0014] The beneficial effects of this invention are as follows: Utilizing UWB centimeter-level positioning, it accurately captures the microscopic behavior of vehicles before the stop line, significantly improving the accuracy of timing estimation; it reduces implementation costs by eliminating the need to modify traffic signals or deploy additional detection equipment; it is applicable to fixed-timing, inductively controlled, or adaptive signalized intersections; as vehicle data continues to flow in, the timing estimation results can be dynamically corrected; by grouping processing by approach lanes, it can be extended to complex multi-phase intersections; by introducing a conflict loss function based on behavioral consistency and boundary tolerance processing, it effectively suppresses timing misjudgments caused by right-turning vehicles, abnormal driving, or data noise. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the signal intersection timing parameter estimation method based on UWB high-precision positioning data described in this invention. Detailed Implementation

[0016] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0017] Numerous specific details are set forth in the following description to provide a thorough understanding of the invention. The described embodiments are only a part of the embodiments of the invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0019] Example 1, as Figure 1 As shown, a method for estimating signal intersection timing parameters based on UWB high-precision positioning data is described, the method comprising: By deploying a UWB base station network in the intersection area, high-precision spatiotemporal trajectory data of multiple passing vehicles is obtained. The high-precision spatiotemporal trajectory data includes timestamp t, two-dimensional coordinates (x,y) and instantaneous speed v. Based on the high-precision spatiotemporal trajectory data and the preset geographic coordinates of the stop lines of each entrance lane, the time and time-speed change curve of each passing vehicle crossing the stop line are obtained. Based on the time-speed change curve, the vehicle traffic pattern is determined, and based on the vehicle traffic pattern, the stopping and starting time and the passing time of each passing vehicle are extracted respectively; The stopping and starting times and passing times of all passing vehicles are merged to construct a passing event sequence that includes both stopping and starting times and passing times; The passage event sequence is subjected to time interval analysis to preliminarily divide the signal period; Under the initially defined signal period, the time in the passage event sequence is mapped to the interval [0, C), where C represents the initially defined signal period; Clustering algorithms are used to cluster the time sequences of traffic events mapped to the signal period, identifying densely populated traffic event intervals as the green light phase window. , ),in, Indicates the initial offset of the green light. Indicates the effective green light duration; A signal timing model is constructed based on the initially defined signal period, green light start offset, and effective green light duration. The timing parameters in the signal timing model are then fine-tuned by constructing an optimization objective function to obtain the optimal timing parameter solution.

[0020] In specific application scenarios, a UWB positioning base station (Anchor) is deployed on each of the four corner lampposts or traffic light poles at the intersection area to construct a positioning network covering the center of the intersection and a range of 100 meters behind the stop lines of each approach lane. Ride-hailing vehicles or buses equipped with UWB positioning tags are selected as probe vehicles. The UWB positioning accuracy is better than 30 centimeters. The return frequency of the UWB tags is set to 10Hz. The raw data stream format received by the backend server is as follows: , in, To indicate the vehicle code, the coordinates of the stop line of the approach lane are pre-mapped. The stop line is set to the y=0 plane, and the vehicle's direction of travel is along the positive y-axis. For vehicles numbered... For vehicles, calculate their distance d(t) and speed v(t) near the stop line. Pass Mode: If a vehicle is within the range of -20m, 20m from the stop line and its minimum speed vmin > 3m / s (not stopped), it is marked as a passing vehicle. ∈ If a vehicle is within the area [-50m, 0m] before the stop line, and its speed (v) is less than 0.5m / s for a duration greater than 3s, and then it accelerates through, it will be marked as a stopped vehicle. ∈ ,for For vehicles in the passage, the time when their front coordinate y(t) = 0 is calculated using linear interpolation and denoted as the passing time. ,for For vehicles in the dataset, extract the moment when their speed abruptly changes from 0 to a threshold (e.g., 1 m / s), and record this as the start-up moment. All data collected during the peak period from 08:00 to 09:00 on the same day will be... and Merge and arrange in chronological order to form a set of candidate green light events. Calculate the time difference between adjacent vehicles in the sequence. A significant long gap was found (caused by red light blocking). Autocorrelation analysis or Fourier transform was used to detect significant frequencies in the event sequence, identifying the period corresponding to the main peak, thus initially dividing the signal period. Using this initial period, all timestamps were mapped to the corresponding periods. ,right One-dimensional density clustering (DBSCAN) was performed, and the results showed that the vehicles passing through the mapped time axis interval [15s, 65s] were the most densely packed and included all of them. Data. Set the initial value of the green light start offset. The initial effective green light duration is g0 = 65 - 15 = 50. At this time, the initial parameter vector is θ0 = (90, 15, 50). The signal timing model is used to characterize the phase state of the traffic lights under a given period C, initial offset φ, and green light duration g.

[0021] One embodiment of the present invention involves obtaining the time and time-speed variation curve of each vehicle crossing the stop line based on the high-precision spatiotemporal trajectory data and the preset geographical coordinates of the stop lines at each approach lane, including: Preset the geographical coordinates of the stop lines at each entrance lane; Using timestamp t as the independent variable, cubic spline interpolation is performed on the discrete two-dimensional coordinates (x,y) of each passing vehicle to construct parametric equations x(t) and y(t) of the position of each passing vehicle with respect to time. The first derivatives of the parametric equations are taken to obtain the component velocities, and the continuous time-velocity variation curves of each passing vehicle are synthesized. And based on the parametric equation, the moment when the coordinates of each passing vehicle coincide with the coordinates of the stop line is obtained.

[0022] In specific application scenarios, the original UWB data is a discrete set of points (one point every 0.1 seconds). Using a cubic spline interpolation algorithm, a cubic polynomial is constructed between adjacent sampling points. Since vehicle motion has physical inertia, its position, velocity, and acceleration should change continuously. The cubic spline not only ensures that the curve passes through all observation points but also guarantees the continuity of the first derivative (velocity) and second derivative (acceleration) at the connection points, best reflecting the kinematic characteristics of a real vehicle. Using the constructed position parameter equations x(t) and y(t), the instantaneous velocity components at any given time can be directly calculated through analytical differentiation. and Then, a continuous velocity curve is obtained through vector synthesis formulas, instead of the traditional finite difference method (Δx / Δt). Geometric intersection solves the vehicle's motion equations and the geometric equations of the stop line simultaneously. The system no longer finds the "nearest point" from a finite number of sampling points, but calculates the precise theoretical moment when the vehicle crosses the coordinate plane of the stop line.

[0023] In one embodiment of the present invention, vehicle traffic patterns are determined based on the time-speed variation curve, and the stopping / starting time and passing time of each passing vehicle are extracted based on the vehicle traffic patterns, including: Based on the time-speed change curve, determine whether the speed of the passing vehicle drops below a preset speed threshold before the stop line and remains above a preset time threshold. If so, mark it as a red light stop, record the time from its stationary state to its acceleration, and define that time as the stop-start time of the passing vehicle. Otherwise, mark the vehicle as a green light through vehicle and record the time it crosses the stop line.

[0024] In one embodiment of the present invention, the clustering algorithm includes the DBSCAN clustering algorithm and the K-means clustering algorithm.

[0025] In one embodiment of the present invention, the optimization objective function is constructed based on a conflict loss function, which is obtained based on the following rules: A high-weight loss occurs when a vehicle passes through a stop line without stopping and the time of its crossing is within the red light period of the signal timing model, or when a vehicle stops but the time of its stop-start is within the red light period of the signal timing model; otherwise, a low-weight loss occurs.

[0026] In specific application scenarios, based on the signal timing parameters of the current iteration (period C, green light offset ϕ, green light duration g), hypothetical red light zones and hypothetical green light zones are divided on the time axis, assuming that all collected data on passing vehicles are driving legally. Therefore, if the model determines that the current time is red, but the observation data shows that vehicles are crossing the stop line or starting, this inconsistency between the model prediction (red) and the actual observation (movement) is defined as a logical conflict. High-weight penalty zone (red light pass / red light start): The loss function is designed to have extremely high values ​​or gradients in the red light interval. Once the parameter θ is set so that the vehicle action falls in the red light zone, the objective function value will increase sharply. This forms a high wall in the mathematical optimization process, forcing the optimization algorithm (such as gradient descent) to quickly adjust the parameters, move the green light window and cover the time points of these vehicle actions; Low-weight tolerance zone (green light pass / green light start): When the vehicle action falls in the green light interval, it is considered a normal phenomenon, and the loss function is assigned an extremely low value. This allows the parameters to fluctuate slightly within a reasonable range to find the optimal solution.

[0027] In one embodiment of the present invention, the specific form of the conflict loss function is as follows: in, This represents the function value of the conflict loss function. Indicates the first The loss caused by a green light through bus. Indicates the first The loss caused by a vehicle stopping at a red light. Indicates the weighting coefficient. Indicates the first The weight of a green light express lane Indicates the first The weight of a vehicle stopping at a red light. This indicates the group of vehicles that have been designated as "green light" through traffic. This indicates the group of vehicles that have been deemed to be stopped at a red light.

[0028] In specific application scenarios ,in, This indicates the maximum acceleration value of the vehicle during the startup process. This represents the average maximum acceleration of all vehicles during startup within a preset sliding time window. The standard deviation of the maximum acceleration during startup. Maximum acceleration during startup. It is the driver's most direct response to the green light switching. The exponent term in the formula is 1. With a denominator of 2 and a final weight of 0.5, the car's start-up is average, providing a reference with moderate reliability; if The exponent approaches 0, the denominator approaches 1, and the final weight is 1. This car starts quickly and is a near-perfect sample (possibly the first car in the queue and fully focused). If The final weight is approximately 0.12. This car starts very slowly (e.g., when playing on a mobile phone or when a novice driver), with a large lag in start-up time, and therefore has a low weight. This allows for a more accurate search for the optimal solution and improves data quality.

[0029] in, This indicates the instantaneous speed of the vehicle as it crosses the stop line. Indicates the road speed limit. This represents the absolute value of the average acceleration during the crossing process. This indicates the preset acceleration threshold. This represents the driver's confidence in passing. A high value indicates that the driver has a clear green light in their field of vision and is willing to pass at full speed; a low value indicates that the driver is hesitant (perhaps seeing that the green light is about to end, or that the road conditions are complicated), and this data has less reference value. This represents the absolute value of the average acceleration over a period of time before and after the vehicle crosses the stop line (e.g., 1 second before and after), calculated using the second derivative of the UWB positioning data. The absolute value is used because a large positive acceleration indicates rapid acceleration (perhaps running a yellow light or just starting out), not a steady-state green light flow; a large negative acceleration (rapid deceleration) indicates braking (perhaps hesitation about stopping, or an obstacle ahead), and is considered interference data. The indicator shows the degree of instability or hesitation in traffic flow. The closer it is to 0, the more it indicates that the vehicle is passing through at a constant speed, which is the most standard behavior for passing through a green light.

[0030] One embodiment of the present invention, in, A function indicating the green light for through traffic; = in, Function to indicate a green light when stopping at a red light in, Indicates the first The time when a green-light through vehicle crosses the stop line; in, Indicates the first The moment a vehicle stops and starts moving when stopped at a red light.

[0031] In one embodiment of the present invention, the green light phase window is composed of the following three consecutive time regions. Green light start blur zone Clearly define the green zone The blurry area ends with the green light The green light initial ambiguity zone is designed to tolerate vehicle start-up delays; vehicles should be within the clearly defined green light zone. Normal passage, the green light ends the ambiguous area. This is to allow vehicles to pass through at the end of the green light period.

[0032] In one embodiment of the present invention, the optimization objective function is: in, This represents the optimal timing parameter estimation solution.

[0033] Construct an optimization objective function, and continuously adjust the timing parameters θ (cycle, green light ratio, phase difference) until all vehicles achieve optimal performance. and All of them should fall as perfectly as possible within the calculated green light window, at which point the total loss is minimized, and the result is... That is the actual signal timing.

[0034] Example 2: A signal intersection timing parameter estimation system based on UWB high-precision positioning data, the system comprising: The spatiotemporal trajectory data acquisition module is used to acquire high-precision spatiotemporal trajectory data of multiple passing vehicles through a UWB base station network deployed in the intersection area. The high-precision spatiotemporal trajectory data includes timestamp t, two-dimensional coordinates (x, y) and instantaneous speed v. The module for acquiring time-velocity change curves acquires the time and time-velocity change curve of each vehicle crossing the stop line based on the high-precision spatiotemporal trajectory data and the preset geographic coordinates of the stop lines of each entrance lane. The vehicle traffic mode determination module determines the vehicle traffic mode based on the time-speed change curve, and extracts the stopping and starting time and passing time of each passing vehicle based on the vehicle traffic mode; A passage event sequence module is constructed to merge the stop start time and passage time of all passing vehicles and construct a passage event sequence containing the stop start time and passage time; A preliminary signal period segmentation module is used to perform time interval analysis on the passage event sequence in order to initially segment the signal period. The time mapping module is used to map the time in the passage event sequence to the interval [0, C) under the initially divided signal period; The clustering module is used to cluster the times in the sequence of traffic events mapped to the signal period using a clustering algorithm, identify dense intervals of traffic events, and use them as the green light phase window. , ),in, Indicates the initial offset of the green light. Indicates the effective green light duration; The optimal solution module is used to construct a signal timing model defined by the initially divided signal period, green light start offset, and effective green light duration, and to fine-tune the timing parameters in the signal timing model by constructing an optimization objective function to obtain the optimal timing parameter solution.

[0035] The embodiments of the present invention are based on the same inventive concept as Embodiment 1 and have the same technical effects, which will not be repeated here.

[0036] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0037] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0038] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0039] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0040] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for estimating timing parameters of signalized intersections based on UWB high-precision positioning data, characterized in that, The method includes: By deploying a UWB base station network in the intersection area, high-precision spatiotemporal trajectory data of multiple passing vehicles is obtained. The high-precision spatiotemporal trajectory data includes timestamp t, two-dimensional coordinates (x,y) and instantaneous speed v. Based on the high-precision spatiotemporal trajectory data and the preset geographic coordinates of the stop lines of each entrance lane, the time and time-speed change curve of each passing vehicle crossing the stop line are obtained. Based on the time-speed change curve, the vehicle traffic pattern is determined, and based on the vehicle traffic pattern, the stopping and starting time and the passing time of each passing vehicle are extracted respectively; The stopping and starting times and passing times of all passing vehicles are merged to construct a passing event sequence that includes both stopping and starting times and passing times; The passage event sequence is subjected to time interval analysis to preliminarily divide the signal period; Under the initially defined signal period, the time in the passage event sequence is mapped to the interval [0, C), where C represents the initially defined signal period; Clustering algorithms are used to cluster the time sequences of traffic events mapped to the signal period, identifying densely populated traffic event intervals as the green light phase window. , ),in, Indicates the initial offset of the green light. Indicates the effective green light duration; A signal timing model is constructed based on the initially defined signal period, green light start offset, and effective green light duration. The timing parameters in the signal timing model are then fine-tuned by constructing an optimization objective function to obtain the optimal timing parameter solution.

2. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 1, characterized in that, Based on the high-precision spatiotemporal trajectory data and the preset geographical coordinates of the stop lines at each approach lane, the time and time-speed variation curve of each vehicle crossing the stop line are obtained, including: Preset the geographical coordinates of the stop lines at each entrance lane; Using timestamp t as the independent variable, cubic spline interpolation is performed on the discrete two-dimensional coordinates (x,y) of each passing vehicle to construct parametric equations x(t) and y(t) of the position of each passing vehicle with respect to time. The first derivatives of the parametric equations are taken to obtain the component velocities, and the continuous time-velocity variation curves of each passing vehicle are synthesized. And based on the parametric equation, the moment when the coordinates of each passing vehicle coincide with the coordinates of the stop line is obtained.

3. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 1, characterized in that, Based on the time-speed variation curve, the vehicle traffic pattern is determined, and based on the vehicle traffic pattern, the stopping and starting time and the passage time of each passing vehicle are extracted, including: Based on the time-speed change curve, determine whether the speed of the passing vehicle drops below a preset speed threshold before the stop line and remains above a preset time threshold. If so, mark it as a red light stop, record the time from its stationary state to its acceleration, and define that time as the stop-start time of the passing vehicle. Otherwise, mark the vehicle as a green light through vehicle and record the time it crosses the stop line.

4. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 1, characterized in that, The clustering algorithms include the DBSCAN clustering algorithm and the K-means clustering algorithm.

5. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 1, characterized in that, The optimization objective function is constructed based on the conflict loss function, which is obtained according to the following rules: A high-weight loss occurs when a vehicle passes through a stop line without stopping and the time of its crossing is within the red light period of the signal timing model, or when a vehicle stops but the time of its stop-start is within the red light period of the signal timing model; otherwise, a low-weight loss occurs.

6. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 5, characterized in that, The specific form of the conflict loss function is as follows: in, This represents the function value of the conflict loss function. Indicates the first The loss value incurred by a green light through bus. Indicates the first The loss caused by a vehicle stopping at a red light. Indicates the weighting coefficient. Indicates the first The weight of a green light express lane Indicates the first The weight of a vehicle stopping at a red light. This indicates the group of vehicles that have been designated as "green light" through traffic. Let θ represent the set of vehicles judged to stop at a red light, and let θ represent the set of timing parameters, θ = {C, , } 7. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 6, characterized in that, in, A function indicating the green light for through traffic; = in, The function that indicates a green light when the traffic light is red. in, Indicates the first The time when a green-light through vehicle crosses the stop line; in, Indicates the first The moment a vehicle stops and starts moving when stopped at a red light.

8. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 1, characterized in that, The green light phase window consists of the following three consecutive time regions. Green light start blur zone Clearly define the green zone The blurry area ends with the green light The green light initial ambiguity zone is designed to tolerate vehicle start-up delays; vehicles should be within the clearly defined green light zone. Normal passage, the green light ends the ambiguous area. This is to allow vehicles to pass through at the end of the green light period.

9. The method for estimating signal intersection timing parameters based on UWB high-precision positioning data according to claim 1, characterized in that, The optimization objective function is: in, This represents the optimal timing parameter estimation solution.

10. A signal intersection timing parameter estimation system based on UWB high-precision positioning data, characterized in that, The system includes: The spatiotemporal trajectory data acquisition module is used to acquire high-precision spatiotemporal trajectory data of multiple passing vehicles through a UWB base station network deployed in the intersection area. The high-precision spatiotemporal trajectory data includes timestamp t, two-dimensional coordinates (x, y) and instantaneous speed v. The module for acquiring time-velocity change curves acquires the time and time-velocity change curve of each vehicle crossing the stop line based on the high-precision spatiotemporal trajectory data and the preset geographic coordinates of the stop lines of each entrance lane. The vehicle traffic mode determination module determines the vehicle traffic mode based on the time-speed change curve, and extracts the stopping and starting time and passing time of each passing vehicle based on the vehicle traffic mode; A passage event sequence module is constructed to merge the stop start time and passage time of all passing vehicles and construct a passage event sequence containing the stop start time and passage time; A preliminary signal period segmentation module is used to perform time interval analysis on the passage event sequence in order to initially segment the signal period. The time mapping module is used to map the time in the passage event sequence to the interval [0, C) under the initially divided signal period; The clustering module is used to cluster the times in the sequence of traffic events mapped to the signal period using a clustering algorithm, identify dense intervals of traffic events, and use them as the green light phase window. , ),in, Indicates the initial offset of the green light. Indicates the effective green light duration; The optimal solution module is used to construct a signal timing model defined by the initially divided signal period, green light start offset, and effective green light duration, and to fine-tune the timing parameters in the signal timing model by constructing an optimization objective function to obtain the optimal timing parameter solution.

Images