CN-122024490-A - Traffic signal lamp intelligent remote control method and system based on multi-source data fusion

CN122024490ACN 122024490 ACN122024490 ACN 122024490ACN-122024490-A

Abstract

The invention relates to the technical field of traffic signal time control, and discloses an intelligent remote control method and system for traffic signal lamps based on multi-source data fusion, wherein the method collects traffic flow data of intersections in real time, and completes space-time alignment of multi-source heterogeneous data by establishing a unified Cartesian coordinate system and an event triggering snapshot mechanism; the method comprises the steps of calculating basic green light time according to confidence weight of vision and radar data and combining safety redundancy coefficient, introducing a downstream back pressure suppression factor item and a historical residual feedback compensation item to correct the basic green light time on the basis of the basic green light time, finally generating a control instruction containing a life cycle time stamp, executing after verifying timeliness of the instruction of the mobile intelligent traffic signal lamp, and uploading actual execution data to update a next cycle scheme. The invention introduces a downstream road saturation monitoring mechanism, solves the problem that the traditional traffic signal lamp control strategy ignores downstream bearing capacity, and improves the robustness of traffic control and the regional cooperative efficiency.

Inventors

REN LINGZHI
ZHOU XIANQI
LIU ZIYANG
WANG QIHUA
Cheng Zengchang
HONG GUOSHENG
LI DINGKUN
XU JUN
GAO JIAN
ZHANG HAILI
LI YANYAN

Assignees

巢湖学院
安徽路顺交通工程有限公司

Dates

Publication Date: 20260512
Application Date: 20260408

Claims (10)

1. A traffic signal lamp intelligent remote control method based on multi-source data fusion is characterized by comprising the following steps: acquiring traffic flow data of an intersection in real time through an intersection sensor network and an internet data source; Establishing an intersection unified Cartesian coordinate system and an event triggering type snapshot mechanism, executing space-time alignment processing, mapping the traffic flow data to the same intersection physical model in space, and synchronizing in time by taking the starting time of a signal control period as a reference; Based on traffic flow data processed by space-time alignment, calculating basic green light time length meeting current crossing traffic demands by utilizing a visual credibility weight coefficient and a radar credibility weight coefficient and combining a lane saturated dissipation speed and maximum traffic capacity; Introducing a downstream back pressure suppression factor item and a historical residual feedback compensation item, constructing a time length correction model, correcting the basic green light time length, and generating corrected green light execution time length; the cloud server packages the corrected green light execution time length into a control instruction with a life cycle time stamp and sends the control instruction to the mobile intelligent traffic signal lamp, the mobile intelligent traffic signal lamp is executed after the verification instruction is time-efficient, and residual queuing data are collected and returned to the cloud after the cycle is finished, so that a scheme of the next cycle is corrected.
2. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 1, wherein the traffic flow data comprises real-time queuing lengths of lanes Dynamic arrival rate of vehicle Vehicle drive-off speed Real-time saturation of downstream road segments ; The real-time queuing length of each lane Acquiring a monitoring video key frame through a high-definition camera, constructing an inverse perspective transformation matrix, mapping image pixel coordinates to road surface world coordinates, and identifying Euclidean distance between the tail part of the vehicle at the farthest end and a parking line in a red light waiting state; The dynamic arrival rate of the vehicle Tracking a moving target track entering a view field through a millimeter wave radar, counting the number of vehicles with speed vectors pointing to an intersection in a set time window, and calculating the average value of the instantaneous speeds of a green light release phase crossing a parking line target to obtain the vehicle driving-out speed ; Real-time saturation of the downstream road segment Saturation by separately obtaining congestion delay index conversions for internet map services And saturation calculated by geomagnetic induction coil of downstream intersection Then take the maximum value of the two to obtain 。
3. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 2, wherein the time-space alignment process comprises projecting the visually recognized vehicle coordinates and the radar target coordinates transformed by the rotation translation matrix to a unified cartesian coordinate system by using a homography matrix and performing spatial position matching, controlling the period by signals Is the starting time of (2) Opening a time window for reference Taking the arithmetic average of radar and vision data in the window and correlating the distances Last time effective internet data For ensuring synchronicity of traffic flow data in the time dimension.
4. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 2, wherein the calculation formula of the basic green light duration is: ; In the formula, Indicating the control period that is currently to be executed, As the base green light duration is used, For the speed of dissipation of the lane saturation, For the maximum traffic capacity of the lane, As the visual credibility weighting coefficient, As the radar confidence weighting coefficient, As a basis for the safety redundancy factor, To be in accordance with the current control period Historical average green light time duration with the same date attribute and in the same statistical window.
5. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 4, wherein the lane saturation dissipation speed is And updating by adopting an exponential weighted moving average algorithm, wherein an updating formula is as follows: ; In the formula, For the vehicle drive-off speed of the current cycle, In order to update the coefficients of the coefficients, Is the minimum flow rate threshold.
6. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 4, wherein the visual credibility weighting coefficient is The radar confidence weight coefficient is the arithmetic average value of the confidence scores of the recognized queuing vehicles in the current period Determining coefficients for linear regression of distance and time track of radar tracking target in current period Root mean square value of (c).
7. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 4, wherein the expression of the duration correction model is: ; In the formula, For the modified green light execution duration, For the downstream back pressure inhibition factor term, As a sensitivity coefficient of the sensor array, In order to trigger a threshold value for congestion, For the historical residual feedback compensation term, For the feedback gain factor to be a function of the gain factor, For the remaining queue length at the end of the last cycle.
8. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 7, wherein verifying the command timeliness comprises the cloud server generating a message containing a command lifecycle timestamp And an effective window After receiving the instruction, the mobile intelligent traffic signal lamp checks the current time and the current time If the difference is smaller than Executing the calculated green light execution time If the difference is greater than or equal to The instruction is discarded and a preset minimum green time is executed.
9. The intelligent remote control method for traffic lights based on multi-source data fusion according to claim 1, wherein the scheme for correcting the next period comprises the steps of calling a vision sensor to solve the residual queuing length at the end time of the current period of the mobile intelligent traffic lights Recording the green light time of actual execution And uploading the historical residual feedback compensation term to the cloud end, and updating the historical residual feedback compensation term of the next period duration correction model.
10. A traffic light intelligent remote control system based on multi-source data fusion, which is used for executing the traffic light intelligent remote control method based on multi-source data fusion as claimed in any one of claims 1 to 9, and is characterized by comprising the following steps: the multi-source sensing module is used for acquiring real-time queuing length, dynamic arrival rate of vehicles, vehicle driving-off speed and real-time saturation of downstream road sections of each lane through an intersection sensor network and an internet data source; The space-time alignment module is used for establishing an intersection unified coordinate system and an event triggering type snapshot mechanism, and carrying out space position mapping and time reference synchronization on the collected heterogeneous traffic flow data so as to realize multi-source fusion of the data; the intelligent calculation module is used for calculating the basic green light time length according to the fused data, constructing a time length correction model by combining the downstream back pressure suppression factor item and the historical residual feedback compensation item, and dynamically generating the final green light execution time length; And the remote control module is used for generating and issuing a control instruction containing a life cycle time stamp, driving the signal lamp to execute after verifying the timeliness of the instruction, and collecting residual queuing data after the execution to perform closed-loop feedback.

Description

Traffic signal lamp intelligent remote control method and system based on multi-source data fusion Technical Field The invention relates to the technical field of traffic signal time control, in particular to an intelligent remote control method and system for a traffic signal lamp based on multi-source data fusion. Background The mobile intelligent traffic signal lamp is used as flexible and efficient temporary traffic control equipment, and the mobile traffic signal lamp is required to perform traffic control temporarily when the original (fixedly installed) traffic signal lamp at the intersection cannot be normally used. Compared with a fixed permanent signal lamp, the intelligent emergency evacuation system has the advantages of self power supply (solar energy/storage battery), no need of laying underground cables, quick deployment and the like, and is important supplementary equipment for realizing regional traffic emergency evacuation and order maintenance of a command center. In the prior art, the control strategy of the mobile traffic signal lamp is relatively lagged, and mainly depends on simple local timing control or manual field operation. Specifically, the existing control modes mainly comprise a preset fixed traffic light countdown scheme for circulation execution and a manual remote controller for on-site use by a traffic police, and the phase is manually switched according to the visual traffic flow condition. The control mode can maintain basic passing order in a regular period of stable traffic flow. However, the conventional mobile signal lamp control method has some limitations in practical application, and is difficult to cope with complex and changeable dynamic traffic flows, and is particularly characterized in that the conventional fixed timing scheme is relatively dead and cannot adapt to the characteristic of severe fluctuation of traffic flow under temporary road conditions, and the control strategy of the next period cannot be dynamically adjusted according to the actual traffic flow change after execution, so that the phenomenon of low efficiency of no-vehicle-free space or vehicle queuing occurs. Secondly, the human remote control mode relies on human eyes to judge that data support is lacking, and fatigue errors are easy to generate. In addition, the mobile signal lamp usually operates as an information island, lacks the capability of coordinating the states of surrounding roads, namely only pays attention to the release of the current intersection, ignores the bearing capability of a downstream road section, and often has the conditions that if the downstream intersection is severely blocked, even if the green light of the current intersection is on, vehicles cannot pass due to the front congestion and even are blocked in the middle of the intersection, so that the traffic jam is caused, or the green light time is too short, so that backlog vehicles are more and more, and finally, the traffic paralysis of the whole area is caused. Therefore, there is a need for a traffic signal intelligent remote control method and system based on multi-source data fusion to solve the above problems. Disclosure of Invention Aiming at the problems in the related art, the invention provides an intelligent remote control method for traffic signal lamps based on multi-source data fusion, which aims to overcome the technical problems in the prior art. In order to solve the technical problems, the invention is realized by the following technical scheme: The embodiment of the invention provides an intelligent remote control method of a traffic signal lamp based on multi-source data fusion, which specifically comprises the steps of collecting traffic flow data of an intersection in real time through an intersection sensor network and an Internet data source, establishing an intersection unified Cartesian coordinate system and an event triggering type snapshot mechanism, executing time-space alignment processing, mapping the traffic flow data to a same intersection physical model in space, synchronizing the time with the starting time of a signal control period as a reference, calculating a basic green light time length meeting the current intersection traffic demand by utilizing a visual credibility weight coefficient and a radar credibility weight coefficient based on the traffic flow data of the time-space alignment processing, combining a lane saturated dissipation speed and the maximum traffic capacity, constructing a time length correction model, correcting the basic green light time length, generating corrected green light execution time length, enabling the downstream back pressure suppression factor item to be used for reducing cloud end time when the downstream congestion occurs, enabling the history residual feedback compensation item to be used for compensating the residual of the previous period, enabling the corrected green light execution time to be packaged into the intelligent green l