Operational evaluation method for signalized intersections based on radar vision fusion detection

Abstract

To provide an operational evaluation method for signalized intersections based on radar vision fusion detection.SOLUTION: A radar vision fusion all-in-one machine is installed at a signalized intersection for real-time detection of the intersection. A detection area is divided into a grid pattern based on a flow path. The number of vehicles in queue, the number of vehicles crossing a stop line, and the number of vehicles passing through the stop line within each time interval are tabulated, and a delay time, a secondary stop rate, a queue length, and a grid occupancy rate are calculated. Hourly and daily operational evaluation indices using a Topsis evaluation model based on an entropy-weighting method is calculated to evaluate congestion conditions at five different levels. Continuous and accurate vehicle dynamic spatial data is provided through the radar vision fusion all-in-one machine. In addition, the Topsis evaluation model based on the entropy-weighting method is used to quantify the evaluation indices and improve the objectivity and accuracy of the evaluation.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the technology of intelligent traffic detection, and more particularly to a method for evaluating the operation of signalized intersections based on radar vision fusion detection. [Background technology]

[0002] Radar vision fusion detection technology is widely used in areas such as traffic information sensing and traffic target detection. Currently, all-in-one radar vision fusion machines integrate emerging technologies such as active forward scanning and millimeter-wave 3DHD detection to achieve wide-area target tracking, high-precision distance measurement, continuous speed measurement, and multi-dimensional directional detection in traffic scenarios. Furthermore, radar vision fusion maximizes the characteristics of composite sensors, converting single-point detection into multi-point joint control, effectively solving the technical problem of large detection deviations caused by single radar or video detection due to interference from shielding or harsh environments. In this regard, radar vision fusion technology has clear advantages in the operational evaluation of signalized intersections, and through joint calibration methods, it corrects various basic vehicle data collected by different devices within a certain range of the intersection, ensuring the accuracy and effectiveness of signalized intersections.

[0003] At this stage, the data types used for operational evaluation of signalized intersections are generally classified into trajectory data and speed data. Trajectory data is mainly divided into GPS trajectories of moving vehicles and GPS trajectories of floating vehicles, while speed data is further divided into vehicle speed and road section speed. From the spatiotemporal dimension analysis obtained from the above data, it was found that the current signalized intersection evaluation data has the following shortcomings: (1) In the time dimension, periodic data collection is insufficient, resulting in low temporal continuity and reliance on past statistical data, thus lacking real-time capabilities. (2) In the spatial dimension, static drift occurs in the target trajectory, meaning there is a large deviation between GPS positioning and the actual trajectory. This necessitates pre-processing of the data to adjust for drift points, increasing the time cost of evaluation.

[0004] Mathematical methods such as hierarchical analysis, fuzzy logic, Gray correlation, and entropy weighting have been used to establish evaluation metrics for intersection operations, and then to categorize the operational status of intersections into different levels to distinguish between smooth and congested states. However, many of these methods suffer from problems such as subjective analysis, ambiguous definitions, and qualitative evaluation, raising questions about the reliability of the evaluation metrics. [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] The objective of the present invention is to overcome the shortcomings of the prior art and provide a method for evaluating the operation of signalized intersections based on radar vision fusion detection. This method selects four key evaluation parameters—total intersection delay, secondary stop rate, queue length, and grid occupancy rate—through a Topsis evaluation model, calculates hourly and daily operational evaluation indicators for the intersection, and avoids the influence of a single indicator and subjective factors on evaluation accuracy. Furthermore, the evaluation method of the present invention is more convincing than qualitative evaluations because it dynamically selects key evaluation indicators using primary and secondary factor analysis and assigns specific values ​​to each key indicator. [Means for solving the problem]

[0006] A method for evaluating the operation of signalized intersections based on radar vision fusion detection, Step S1 involves installing a mature radar vision fusion all-in-one machine at an intersection traffic light, setting the rated operating parameters, and detecting static information of the intersection and dynamic information of vehicles in real time. Step S2 constructs a digital twin scene of the intersection, defines a rectangular detection area from the stop line of each entry lane to the maximum range of radar vision fusion detection, divides it into small grids using the same vehicle length as the unit, and gives the coordinates of the intersections of the small grids. Step S3 involves setting detection time intervals based on the stop line, aggregating the number of vehicles in the queue within each time interval, the number of vehicles that stop and start beyond the stop line, and the number of vehicles that stop and start without crossing the stop line, and calculating the total intersection delay, secondary stopping rate, queue length, and grid occupancy rate. The Topsis evaluation model uses a scheme index a for each evaluation period Ai, with evaluation periods set to every 15 minutes per hour or every hour per day, and uses total intersection delay, secondary stop rate, queue length, and grid occupancy rate. ij Step S4 calculates hourly and daily operational evaluation indicators and evaluates the congestion status of signalized intersections by dividing them into five levels. The above method, including.

[0007] Preferably, step S1 specifically includes the following steps: Step S11 defines the rated operating parameters of the radar vision fusion all-in-one machine, the maximum scanning distance D of the radar vision fusion all-in-one machine, the scanning frequency f, and the vehicle resolution ε. Static information of the intersection: t number of lanes at the mth intersection entrance m The width of the nth lane of the entrance ramp to the mth intersection is W. m,n Step S12 defines as follows: The dynamic information of the vehicles includes the 2D coordinates of the kth detected vehicle, vehicle classification (labeling small passenger cars as "vehicle," medium-sized cars as "midvehicle," and large cars as "bigvehicle"), and the real-time speed of the kth vehicle. k Real-time azimuth angle φ of the kth vehicle k Step S13 defines as, S14 involves co-calibrating the 2D coordinates of the detected vehicle using coordinate radar vision fusion.

[0008]

number

[0009] Preferably, the construction of the digital twin scene of the intersection in step S2 is Specifically: Define the coordinates of the intersections of the small grids, use the stop line as the baseline of the rectangular area, and use the joint calibration method in step 1 to obtain the coordinates (u 0,m , v 0,m ) of the intersections of the stop line and each incoming lane. u 0,m is the horizontal coordinate of the intersection of the m-th lane and the stop line, v 0,m is the vertical coordinate of the intersection of the m-th lane and the stop line. Define the coordinates of the remaining intersections as (u r,m , v r,m ), where r is the r-th small grid up to the maximum detection range direction based on the stop line, and define the lane closest to the U-axis as the first lane (m = 1). Step S21 is included.

[0010] Preferably, step S3 specifically includes the following steps.

[0011] S31: The step of calculating the intersection delay.

[0012]

Number

[0013] The method for detecting the number of stopped vehicles S i and the number of non-stopped vehicles nS i is as follows: Taking the stop line coordinates as the detection reference, within one signal cycle, for a vehicle coordinates (uk ,v k ) exceeds the stop line coordinates or the velocity is from 0 to v k When it changed to S i This is accumulated once; a certain vehicle coordinate (u k ,v k ) exceeds and the speed is v k from v k1 When it changed to nS i This is accumulated once, S32: Step to calculate the secondary shutdown rate.

[0014]

number

[0015] S33: Step to calculate the queue length.

[0016]

number

[0017] S34: Step to calculate grid occupancy.

[0018]

number

[0019] Preferably, the calculation steps for the hourly performance indicator and the daily performance indicator Topsis in step S4 are as follows:

[0020] S41: Standardization step, Prospective index r ij It can be expressed by the following formula.

[0021]

number

[0022] S42: Attribute Weight W j Steps to determine (1) The entropy value EH of the j-th index in the entropy weighting method j Calculate.

[0023]

number

[0024] (2) Time difference coefficient FH using entropy weighting method j Calculate.

[0025] (Math 8) FH j = 1 - EH j , j=1,2,3,4

[0026] (3) Time-based operational evaluation weights WH using entropy weighting method j To decide.

[0027]

number

[0028] Similarly, heavy WD for daily operational evaluation j That will also be decided.

[0029]

number

[0030] S43: Step to calculate the weighted norm matrix A.

[0031]

number

[0032] S44: A step to determine the ideal positive and ideal negative solutions for matrix A.

[0033]

number

[0034] S45: A step to calculate the Euclidean distance between the positive ideal solution and the negative ideal solution in each evaluation period.

[0035]

number

[0036] S46: A step to calculate the relative proximity of Ai to the ideal solution during each evaluation period.

[0037]

number

[0038] ch * i Each evaluation period is normalized from longest to shortest and sorted accordingly, with the evaluation periods ranked higher being considered superior.

[0039]

number

[0040]

number

[0041] Furthermore, the classification criteria for evaluation indicators, namely AH k ∈[0,20%] is very good within 1 hour, AH k ∈[20%,40%] is progressing fairly well within the first hour, AH k ∈[40%,60%] are progressing smoothly within 1 hour, AH k ∈[60%,80%] will experience some congestion within 1 hour, AH k ∈[80%,100%] is defined as very congested for less than one hour. [Effects of the Invention]

[0042] The advantages and technical effects of the present invention are as follows:

[0043] This invention better solves the problem in intersection operation evaluation where data exists in two dimensions of space and time. By using radar vision fusion detection technology, the data has strong temporal continuity and real-time capabilities. By abandoning the conventional GPS coordinate system and establishing a unique coordinate system in the form of grid division, it avoids the processing of static drift points and reduces the computational load of cloud processing. [Brief explanation of the drawing]

[0044] [Figure 1] This is a flowchart of the operational evaluation method for signalized intersections based on radar vision fusion detection according to the present invention. [Figure 2] This figure shows a grid division model of intersection entrance roads constructed based on the intersection digital twin of the present invention. [Modes for carrying out the invention]

[0045] To further understand the content, features, and effects of the present invention, examples are given below and described in detail with reference to the accompanying drawings. Please note that these examples are for illustrative purposes only and are not intended to limit the present invention or the scope of protection of the present invention.

[0046] Taking a city intersection as an example, as shown in Figure 1, the method for evaluating the operation of a signalized intersection based on radar vision fusion detection according to the present invention includes the following steps.

[0047] (1) First, complete the acquisition of basic data and detect static information of intersections and dynamic information of vehicles in real time.

[0048] Let Y1, Y2, Y3, and Y4 be the operational status of the intersection evaluated every 15 minutes within an hour, and let X1, X2, X3, and X4 be the four attributes of total intersection delay, secondary stopping rate, queue length, and occupancy rate, respectively, and calculate the results for each every 15 minutes. ij (i=1, 2, 3, 4, j=1, 2, 3, 4) represents the possible values ​​(also called the original weights) of scheme Xi for attribute Yi.

[0049] [Table 1]

[0050] Static information for an intersection is the number of lanes t of the entrance road to the mth intersection. m The width of the nth lane of the entrance ramp to the mth intersection is W. m,n That is

[0051] The dynamic information of the vehicle is the 2D coordinate (u k ,v k ), vehicle classification (labeling small passenger cars as "vehicle," medium-sized cars as "midvehicle," and large cars as "bigvehicle"), real-time speed of the kth vehicle v k Real-time azimuth angle φ of the kth vehicle k That is the case.

[0052] The joint calibration of the vehicle's 2D coordinate radar vision fusion is expressed by the following equation.

[0053]

number

[0054] (ii) Define a rectangular detection area from the stop line of each entrance to the maximum range of radar vision fusion detection, divide it into small grids using the same vehicle length as the unit, and give the coordinates of the intersection of the small grids.

[0055] Definition of coordinates of intersections of small grids: The stop line is used as the baseline of the rectangular area, and the coordinates of the intersections of the stop line and each lane are determined by the joint calibration method of Step 1 (u 0,m ,v 0,m ) decide, u 0,m v is the horizontal coordinate of the intersection of the mth lane and the stop line. 0,m (u) is the vertical coordinate of the intersection of the mth lane and the stop line, and the coordinates of the remaining intersections are (u r,m ,v r,m ) is defined as follows, and r is the r-th small grid from the stop line to the direction of the maximum detection range.

[0056] (iii) A step of calculating the total intersection delay, secondary stop rate, queue length, and grid occupancy rate.

[0057] Specifically, the intersection delay can be expressed by the following formula.

[0058]

number

[0059] Number of stopped vehicles s i and the number of vehicles that did not stop nS i Detection method: Using the stop line coordinates as the detection reference, within one signal cycle, a certain vehicle coordinate (u k ,v k ) exceeds the stop line coordinates or the velocity is from 0 to v k When it changed to S i This is accumulated once; a certain vehicle coordinate (u k ,v k ) exceeds and the speed is v k from v k1 When it changed to nS i This is accumulated once, Specifically, the next stop rate is expressed by the following formula.

[0060]

number

[0061] The queue length in step 3 is expressed by the following formula.

[0062]

number

[0063] The grid occupancy rate in Step 3 is expressed by the following formula.

[0064]

Number

[0065] (4) Step of calculating the operation evaluation indicators by time and daily operation evaluation indicators using the TOPSIS evaluation model and evaluating the congestion status of signalized intersections by dividing them into five levels.

[0066] Step 1: Standardization: The forward index r ij is expressed by the following formula.

[0067]

Number

[0068] Step 2: Determination of the attribute weight W j : (1) Calculate the entropy value EH j of the j-th index by the entropy weight method.

[0069]

Number

[0070] (2) Calculate the time difference coefficient FH j by the entropy weight method.

[0071] <00OO453>(Equation 24) FH j = 1 - EHj , where j = 1, 2, 3, 4

[0072] (3) Determine the weight WH of the time - based operation evaluation by the entropy weighting method j .

[0073]

Number

[0074] Similarly, determine the weight WD of the daily operation evaluation j too.

[0075]

Number

[0076] Step 3: Calculate the weighted norm matrix A.

[0077]

Number

[0078] Step 4: Determine the positive ideal solution and negative ideal solution of matrix A.

[0079]

Number

[0080] Step 5: Calculate the Euclidean distance between the positive ideal solution and the negative ideal solution in each evaluation period.

[0081]

Number

[0082] Step 6: Calculate the relative closeness of Ai to the ideal solution in each evaluation period.

[0083]

Number

[0084] Step 7: ch * i Each evaluation period is normalized from longest to shortest and sorted accordingly, with the evaluation periods ranked higher being considered superior.

[0085]

number

[0086]

number

[0087] Furthermore, the classification criteria for evaluation indicators, namely AH k ∈[0,20%] is very good within 1 hour, AH k ∈[20%,40%] is progressing fairly well within the first hour, AH k ∈[40%,60%] are progressing smoothly within 1 hour, AH k ∈[60%,80%] will experience some congestion within 1 hour, AH k ∈[80%,100%] is defined as very congested for less than one hour.

[0088] Finally, for aspects not described in this invention, mature products and technical means from the prior art are employed.

[0089] Those skilled in the art will understand that various improvements or modifications can be made based on the above description, and that such improvements and modifications are also covered by the claims of the appended patent of the present invention.

Claims

1. A method for evaluating the operation of signalized intersections based on radar vision fusion detection, Step S1 involves installing a radar vision fusion all-in-one machine at an intersection traffic light, setting rated operating parameters related to predetermined operating conditions of the radar vision fusion all-in-one machine, and detecting static information of the intersection and dynamic information of vehicles in real time. Step S2: Based on the above detection, a digital twin scene is constructed that generates a dynamic and real-time virtual model in digital space that perfectly corresponds to the actual intersection, a rectangular detection area is defined from the stop line of each entrance lane to the maximum range of radar vision fusion detection, a small grid is divided using the same vehicle length as the unit, and the coordinates of the intersections of the small grid are given. Step S3 involves setting detection time intervals based on the stop line, aggregating the number of vehicles in the queue within each time interval, the number of vehicles that stop and start beyond the stop line, and the number of vehicles that stop and start without crossing the stop line, and calculating the total delay at the intersection, the secondary stop rate (the rate at which vehicles stop more than once at the same signal), the queue length, and the grid occupancy rate. The Topsis evaluation model uses an evaluation period Ai of 15 minutes per hour or 1 hour per day, and uses scheme indicator a for each evaluation period, measuring total intersection delay, secondary stop rate, queue length, and grid occupancy rate. ｉｊ This includes step S4, which calculates hourly operational evaluation indicators and daily operational evaluation indicators, and evaluates the congestion situation at signalized intersections by dividing them into five levels. Step S3 specifically includes the following steps S31: Step of calculating the total delay at the intersection, [Math 2] [In the formula, N is the total number of vehicles that stopped and started, n ｉ，ｊ ∫ is the number of vehicles stopped at different times on the intersection entrance road within △t at the jth minute, n is the detection time (minutes), m is the number of intervals divided into 1 minute, S is the total number of stopped vehicles, S ｉ ∫ is the number of vehicles stopped on the intersection entrance ramp within i minutes, α is the total delay at the intersection, △t is the detection time interval of less than 1 minute, β is the average delay of each stopped / starting vehicle, t ｍａｘ This is the maximum detection time. Number of stopped vehicles S ｉ and the number of vehicles that did not stop nS ｉ Detection method: Taking the stop line coordinates as the detection criterion, within one signal cycle, when a vehicle coordinate (u ｋ , v ｋ ) exceeds the stop line coordinates or the speed changes from 0 to v ｋ , S ｉ is cumulated once; when a vehicle coordinate (u ｋ , v ｋ ) is exceeded and the speed changes from v ｋ to v ｋ１ , nS ｉ is cumulated once. S32: Step to calculate the secondary shutdown rate, [Math 3] [In the formula, r is the secondary stopping rate for the entire intersection, and ds is the coordinate of a vehicle (u ｋ ,v ｋ ) does not exceed the stop line coordinates, and the velocity state is 0 or v ｋ１ This is the cumulative number of secondary stops when the speed changes and then decelerates further to 0. S33: Step to calculate the queue length, [Math 4] [In the formula, L is the total length of the queue at the intersection, M is the total number of entrances to the intersection, I is the total number of lanes in the entrances to the intersection, l ｍ，ｉ [This is the total length of the queue of waiting vehicles in the i-th lane of the entrance ramp to the m-th intersection.] S34: Step to calculate the total grid occupancy rate. [Math 5] [In the formula, δ is the grid occupancy rate, O is the area occupied by queuing vehicles within the grid, Q is the total area of ​​the grid, M is the total number of intersection entrances, I is the total number of lanes in the intersection entrances, O ｍ，ｉ This represents the grid area occupied by the waiting vehicles in the i-th lane of the entrance ramp to the m-th intersection. The calculation steps for the hourly performance evaluation index and the daily performance evaluation index Topsis in step S4 are as follows: S41: Standardization step, Positive indicators RR ｉｊ It is expressed by the following equation: [Math 6] S42: Attribute weight lol ｊ Steps to determine (1) The entropy value EH of the j-th index in the entropy weighting method ｊ Calculate, [Number 7] (2) Time difference coefficient FH using the entropy weighting method ｊ Calculate, (Math 8) FH ｊ =1-EH ｊ , j=1,2,3,4 (3) Weights WH for time-based operational evaluation using entropy weighting method ｊ We decided, [Number 9] Similarly, the weights of the daily operational evaluation WD ｊ It has also been decided, [Number 10] S43: Step of calculating the weighted norm matrix A, [Math 11] S44: Steps to determine the ideal positive and negative solutions for matrix A. [Math 12] S45: A step to calculate the Euclidean distance between the positive ideal solution and the negative ideal solution in each evaluation period. [Number 13] S46: A step to calculate the relative proximity of Ai to the ideal solution during each evaluation period. [Number 14] ch ＊ ｉ The evaluation periods are normalized from longest to shortest and sorted accordingly, with the higher-ranked evaluation periods being considered superior. [Number 15] Similarly, [Number 16] [In the formula, AH ｋ This is the operational evaluation indicator for the i-th evaluation period of less than one hour, AD. ｋ is the operational evaluation indicator for the i-th evaluation period of one day or less, a ｉ，ｊ is attribute Y ｉ X ｉ These are the possible values. Classification criteria for evaluation indicators, i.e., AH ｋ ∈[0,20%] is very good within 1 hour, AH ｋ ∈[20%, 40%] is progressing fairly well within 1 hour, AH ｋ ∈[40%, 60%] are progressing smoothly within 1 hour, AH ｋ ∈[60%, 80%] indicates moderate congestion within 1 hour, AH ｋ ∈[80%, 100%] is defined as extremely congested for less than one hour. A method for evaluating the operation of a signalized intersection based on radar vision fusion detection, characterized by the above.

2. Step S1 is, Step S11 defines the rated operating parameters of the radar vision fusion all-in-one machine, the maximum scanning distance D of the radar vision fusion all-in-one machine, the scanning frequency f, and the vehicle resolution ε. The static information of the aforementioned intersection is the number of lanes t of the entrance road to the mth intersection. ｍ The width of the nth lane of the entrance ramp to the mth intersection is W. ｍ，ｎ Step S12 defines as follows: The dynamic information of the aforementioned vehicles includes the 2D coordinates of the kth detected vehicle, vehicle classification (labeling small passenger cars as "vehicle," medium-sized cars as "midvehicle," and large cars as "bigvehicle"), and the real-time speed of the kth vehicle. ｋ Real-time azimuth angle φ of the kth vehicle ｋ Step S13 defines the following, S14 Coordinate radar vision fusion is used to jointly calibrate the 2D coordinates of the detected vehicle. A method for evaluating the operation of a signalized intersection based on radar vision fusion detection, as described in claim 1, characterized by including the following: [Math 1] [In the formula, (u ｋ ,v ｋ ) is the 2D pixel coordinate of the vehicle after joint calibration using Zhang's calibration method, (x ｋ , y ｋ ) represents the coordinates after camera distortion correction.

3. The construction of the digital twin scene of the intersection in step S2 is as follows: Specifically, define the coordinates of the intersection points of the small grid, use the stop line as the baseline of the rectangular area, and use the joint calibration method of step 1 to determine the coordinates of the intersection points of the stop line and each on-ramp lane (u ０，ｍ ,v ｏ，ｍ ) decided, u ０，ｍ v is the horizontal coordinate of the intersection of the mth lane and the aforementioned stop line. ｏ，ｍ (u) is the vertical coordinate of the intersection of the mth lane and the stop line, and the coordinates of the remaining intersections are (u) ｒ，ｍ ,v ｒ，ｍ The method for evaluating the operation of a signalized intersection based on radar vision fusion detection according to claim 2, characterized in that it includes step S21, in which r is defined as the r-th small grid from the stop line to the direction of the maximum detection range, and the lane closest to the u-axis is defined as the first lane (m=1).

Images