Signal lamp intelligent control method, system and equipment based on multi-modal data

Abstract

The invention provides a signal lamp intelligent control method, system and device based on multi-modal data, and relates to the field of traffic control, and the method comprises the steps: collecting the traffic data of a monitored road section and the data of a vehicle-mounted terminal in real time; the traffic data comprises the number of vehicles, the number of non-uniform-speed driving times and the driving time; the vehicle-mounted terminal data comprises the fuel consumption accumulated usage amount of the vehicle passing through the monitoring road section, the passing time of the vehicle passing through the monitoring road section and the road section length of the monitoring road section; the monitoring road section comprises a plurality of adjacent intersections; regulating and controlling intersection traffic lights according to the traffic data, and generating a regulation and control scheme; judging whether the regulation and control scheme is an optimal regulation and control scheme or not according to vehicle-mounted terminal data; if yes, regulating and controlling intersection traffic lights according to the regulation and control scheme, and determining an optimal driving route according to the vehicle-mounted terminal data; and if not, determining a regulation and control scheme according to the past data. The traffic light duration can be regulated and controlled, the exhaust emission is reduced, and the traffic jam is reduced.

Description

Technical Field

[0001] This invention relates to the field of traffic control, and in particular to a method, system and device for intelligent control of traffic lights based on multimodal data. Background Technology

[0002] Exhaust emissions are particularly severe in congested areas. Frequent stop-and-go driving puts vehicles in uneconomical conditions, where emissions are 5-10 times higher than during normal, steady-speed driving. This air pollution seriously threatens people's physical and mental health.

[0003] While existing macro-level intersection control measures can alleviate congestion to some extent, relying solely on flow data during rush hours is insufficient to adapt to complex traffic changes. Furthermore, traditional control methods often focus only on "throughput," neglecting the "driving quality" of vehicles during transit (such as acceleration / deceleration frequency and fuel consumption). This can easily lead to situations where, despite extended green light durations, vehicles still need to frequently start and stop, resulting in longer queues at intersections and persistently high exhaust emissions.

[0004] Fuel consumption and exhaust emissions, as direct consequences of congestion, are important indicators for evaluating the effectiveness of traffic control. Optimization algorithms should be based on multimodal data (flow rate + energy consumption + trajectory), utilizing real-time feedback from onboard terminals to close-loop correct traffic light control strategies. This can reduce exhaust emissions, smooth traffic flow, and achieve uniform speeds as much as possible, thereby effectively improving traffic efficiency and reducing environmental pollution. Summary of the Invention

[0005] The purpose of this invention is to provide a traffic light intelligent control method, system and device based on multimodal data, which can combine macro traffic data and micro vehicle energy consumption data to regulate traffic lights, reduce exhaust emissions and reduce traffic congestion.

[0006] To achieve the above objectives, The first aspect of this invention provides a method for intelligent control of traffic lights based on multimodal data, comprising the following steps: Real-time collection of traffic data and vehicle terminal data of the monitored road sections, and determination of the target time period to which the current moment belongs; Based on the traffic data, a traffic congestion index is calculated, and preliminary control plans are made for traffic lights at intersections to generate preset control schemes. Based on the actual traffic efficiency fed back by the vehicle terminal data, determine whether the preset control scheme is the optimal control scheme (i.e., verify the effectiveness of the scheme). If so, the traffic lights at the intersection are controlled according to the preset control scheme, and the optimal driving route is determined based on the data from the vehicle terminal. If not (or if the data deviation is too large), activate the safety mode and determine the control plan based on past data.

[0007] In conjunction with the first aspect, the present invention is further configured such that: the traffic data includes the number of vehicles, the number of times of non-uniform speed travel, and the travel time; the vehicle terminal data includes the real-time cumulative fuel consumption of vehicles passing through the monitored road segment, the passage time of vehicles passing through the monitored road segment, and the length of the monitored road segment; the monitored road segment includes multiple adjacent intersections; The historical data includes the total number of vehicles, average number, average green light duration, and cycle for the target time period over the past week.

[0008] In conjunction with the first aspect, the present invention is further configured to: regulate the traffic lights at the intersection based on the traffic data, and plan the regulation of the traffic lights at the intersection based on the traffic congestion index, generating a preset regulation scheme, including: The first data is calculated using the formula p1=n×t / m; where p1 is the first data (reflecting the degree of road segment disturbance); n is the number of non-uniform speed trips; t is the travel time; and m is the number of vehicles. Filter the first data to obtain the filtered first data; Based on the filtered first data, the traffic lights at the intersection are adjusted to generate a preset adjustment scheme. The control operations include increasing the green light duration, decreasing the green light duration, adjusting the cycle, and controlling the phase difference.

[0009] In conjunction with the first aspect, the present invention is further configured to: determine whether the control scheme is the optimal control scheme based on the data from the vehicle terminal, specifically including: The second data is calculated using the formula p2=t' / s; where p2 is the second data (time per unit distance); t' is the time it takes for a vehicle to pass through the monitored road segment; and s is the length of the road segment. Filter the second data to obtain the filtered second data; If the absolute value of the difference between the filtered second data and the theoretical time corresponding to the expected control effect in the control scheme is within the error range, the control scheme is determined to be the optimal control scheme.

[0010] In conjunction with the first aspect, the present invention is further configured to: determine the optimal driving route based on the vehicle terminal data, specifically including: The third data is calculated using the formula p3=l / s; where p3 is the third data (energy consumption per unit distance); l is the cumulative fuel consumption of the vehicle passing through the monitored road segment; and s is the length of the monitored road segment. Filter the third data to obtain filtered third data; The optimal driving route is determined based on the filtered second data (time cost) and the filtered third data (economic cost).

[0011] In conjunction with the first aspect, the present invention is further configured such that: the actual cumulative fuel consumption of the vehicle during the passage of the monitored road segment is: real-time fuel injection data or instantaneous energy consumption data of the target vehicle is obtained through a data monitoring device (such as an OBD interface) installed on the target vehicle; combined with the real-time GPS location information of the vehicle, the cumulative fuel consumption of the vehicle during the passage of the monitored road segment is calculated by integration.

[0012] A second aspect of the present invention also provides an apparatus / device / system for intelligent control of traffic lights based on multimodal data, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0013] A third aspect of the present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0014] A fourth aspect of the present invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0015] According to specific embodiments provided by the present invention, the following technical effects are disclosed: The embodiments of the present invention construct a closed-loop control system of "macro-flow monitoring - micro-energy consumption feedback" by real-time collection of traffic data from monitored road sections and data from vehicle terminals. A preliminary control plan is generated based on the traffic data, and the effectiveness of the plan is verified using the real traffic efficiency (time and fuel consumption) returned by the vehicle terminals. If the plan is verified as optimal, it is implemented and promoted; if the deviation is too large, it reverts to the previous data mode to ensure the safety of the control. This method can effectively regulate the duration of traffic lights, reduce unnecessary starts and stops and idling of vehicles at intersections, thereby significantly reducing exhaust emissions and alleviating traffic congestion. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 The flowchart of the intelligent traffic light control method based on multimodal data provided by the present invention is shown below; Figure 2The basic control principle diagram provided by this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The purpose of this invention is to provide a method, system, and device for intelligent control of traffic lights based on multimodal data, which reduces exhaust emissions and alleviates traffic congestion.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1: like Figure 1 As shown, this invention provides a smart traffic light control method based on multimodal data, comprising: Step 101: Collect real-time traffic data and vehicle terminal data for the monitored road segment; the traffic data includes the number of vehicles, the number of non-uniform speed driving times, and the driving time; the vehicle terminal data includes the real-time cumulative fuel consumption of vehicles passing through the monitored road segment, the passage time of vehicles through the monitored road segment, and the length of the monitored road segment; the monitored road segment includes multiple adjacent intersections.

[0022] In practical applications, adjacent intersections (approximately 2-10) are divided into a sub-zone, which serves as the monitored road segment, with the intersections experiencing the highest traffic volume being the main intersections. Radar detectors are installed at each intersection to collect real-time traffic flow data, and are wired to a control box (one for each sub-zone). The detectors, control boxes, and vehicle-mounted terminals (such as connected car systems or aftermarket OBD devices) are all wirelessly connected to a cloud platform. The cloud platform collects data from each component, performs multimodal data fusion processing, and then generates control commands.

[0023] The system configures the basic parameter information of the intersection based on the previous data. The basic parameter information includes phase, stage, detector, priority of different control modes, minimum green time, maximum green time, adjustment step size, maximum cycle of the intersection, and setting the phase difference between adjacent intersections.

[0024] The green light start time of the main intersection is used as the reference signal for all intersections in the sub-zone. The phase difference of the main intersection is defined as 0, and the phase difference of other intersections is taken from the data measured or calculated on site.

[0025] Among them, priority is used to select a dominant one when a conflict occurs, and to regulate according to the main intersection.

[0026] Control modes are typically categorized into manual control, central control, and local control, with manual control having the highest priority, followed by central control, and local control having the lowest priority.

[0027] In addition, the priority of each phase at the intersection must be considered. Generally, the phases of the main road have a higher priority, while the phases of the secondary road have a lower priority. The phase configuration aims to minimize the number of times vehicles repeatedly start and stop, thereby minimizing unnecessary exhaust emissions.

[0028] Step 102: Calculate the traffic congestion index based on the traffic data, and plan the control of traffic lights at intersections based on the traffic congestion index to generate a preset control scheme.

[0029] In practical applications, radar detectors are used to detect traffic data (such as cross-sectional flow, queue length, etc.), while the cloud is used to obtain the transit time and energy consumption data uploaded by the vehicle terminal.

[0030] In practical applications, step 102 specifically includes: calculating the first data using the formula p1=n×t / m; where p1 is the first data (road segment disturbance index); n is the number of non-uniform speed driving times (e.g., the number of times the absolute value of acceleration exceeds a certain threshold); t is the cumulative driving time of the vehicle on the monitored road segment; and m is the number of vehicles. The cloud platform filters abnormal data (such as extreme congestion data caused by accidents) from the first data according to a first preset threshold to obtain filtered first data; and adjusts the traffic lights at the intersection according to the filtered first data to generate a preset control scheme; the control operations include increasing the green light duration, decreasing the green light duration, adjusting the cycle, and adjusting the phase difference.

[0031] The control operation sets the detector's judgment logic, phase activation order, and phase priority weights through conditional statements. For example, if p1 is greater than the initially set theoretical optimal value x (indicating large disturbances and frequent vehicle starts and stops), then for every unit increase in deviation, the green light ratio is increased by c seconds or the phase difference is adjusted to smooth traffic flow; if p1 is less than the theoretical optimal value x, then the cycle is appropriately shortened to improve intersection utilization.

[0032] Step 103: Determine whether the control scheme is the optimal control scheme based on the data from the vehicle terminal (i.e., verification step). If yes, proceed to step 104; otherwise, proceed to step 105.

[0033] In practical applications, step 103 specifically includes: calculating the second data using the formula p2=t' / s; where p2 is the second data (reverse speed / time per unit distance); t' is the actual time the vehicle takes to pass through the monitored road segment; and s is the length of the road segment.

[0034] Filter the second data to obtain the filtered second data; The filtering involves setting a value range and filtering out data outside the range (e.g., removing data from vehicles that are parked for business or experiencing malfunctions). The value range is obtained through theoretical calculations and verification based on the passage time of past vehicles through the monitored road segment.

[0035] If the absolute value of the difference between the filtered second data and the expected control effect (theoretical time) of the control scheme is within the error range, the control scheme is determined to be effective and optimal. This means that the scheme calculated by the current traffic flow model matches the actual vehicle operating state.

[0036] Step 104: Adjust the traffic lights at the intersection according to the control scheme, and determine the optimal driving route based on the data from the vehicle terminal.

[0037] In practical applications, step 104 specifically includes: calculating the third data using the formula p3=l / s; where p3 is the third data (energy consumption intensity); l is the cumulative fuel consumption of the vehicle passing through the monitored road segment; and s is the length of the monitored road segment.

[0038] The method for obtaining the actual cumulative fuel consumption *l* of a vehicle passing through a monitored road segment is as follows: The instantaneous fuel injection quantity or battery output power of the target vehicle is obtained through a data monitoring device (such as an OBD / CAN module) installed on the target vehicle. Combined with real-time GPS location information, the cumulative energy consumption of the vehicle during its entry into and exit from the monitored road segment is calculated. Compared to the traditional refueling mileage calculation method, this method offers second-level real-time performance and can accurately reflect the impact of current traffic control on energy consumption.

[0039] The third data is filtered to obtain filtered third data; the optimal driving route is determined based on a comprehensive score of the filtered second data (time) and the filtered third data (energy consumption).

[0040] In practical applications, when filtering the second and third data, data in the second data that exceeds the normal range (such as greater than 5 minutes or less than 30 seconds) is identified as abnormal data and removed.

[0041] Verification logic: Based on the filtered third data, if the third data differs from the theoretical optimal solution calculated by the cloud platform based on the number of vehicles, road length, and traffic light control data (e.g., the vehicle energy consumption should be Y under ideal conditions) by no more than 'a' (error tolerance value), it proves that the above adjustments are effective, and the system maintains the current control strategy and continues to iterate and optimize. Conversely, if the actual energy consumption is much higher than the theoretical value, it proves that the control scheme has deviated (possibly due to sensor failure or a sudden event), and the fallback mechanism is triggered.

[0042] If the control plan is found to be effective, the optimal driving route is determined by combining the second and third data (for example, if the p2 and p3 values ​​of a certain road segment are both the smallest, it means that the road segment is both fast and fuel-efficient). This information is then sent to the vehicle terminal, which connects to the navigation system to notify the relevant vehicles in advance and make further route planning optimizations.

[0043] Step 105: Determine the control plan based on past data; the past data includes the total number of vehicles, average number, average green light duration and cycle for the target time period in the past week.

[0044] The specific steps for determining the control plan based on historical data are as follows: The system retrieves the target time period corresponding to the current moment; it then retrieves historical data for that target time period from the database over the past week, including the total number of vehicles, average green light duration, and cycle; after removing outliers from the historical data, it calculates the arithmetic mean of the remaining data to obtain the baseline green light duration and baseline cycle; and it configures the operating parameters of the intersection's traffic lights to these baseline green light duration and baseline cycle for fixed timing control or as the basis for local sensor control. This step serves as a "safety mode" for the system, ensuring that traffic can still maintain basic operation even if real-time calculations fail or become inaccurate.

[0045] This invention, based on traditional adaptive algorithms, introduces "vehicle terminal energy consumption data" as a core feedback indicator, expanding the judgment criteria from simple congestion conditions (location / speed) to vehicle operating quality (repeated starts / non-uniform speed driving / fuel consumption). A well-configured phase difference allows the vehicle to pass through multiple intersections continuously without stopping, thereby reducing the number of stops, stopping delays, and energy consumption. The basic control principle is as follows: Figure 2 As shown.

[0046] This invention utilizes the automatic adjustment of intersection signal timing by the traffic signal controller and the real-time detection data from the radar probe to maximize the smooth flow of vehicles, minimize repeated starts and non-uniform speed driving, and achieve the goal of reducing pollution.

[0047] Example 2: The present invention also provides a device / equipment / system for intelligent control of traffic lights based on multimodal data, including a memory, a processor and a computer program stored in the memory, characterized in that the processor executes the computer program to implement the steps of the above method.

[0048] The present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0049] The present invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0050] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0051] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for intelligent control of traffic lights based on multimodal data, characterized in that, Includes the following steps: Real-time collection of traffic data and vehicle terminal data of the monitored road sections, and determination of the target time period to which the current moment belongs; The traffic congestion index is calculated based on the traffic data, and the traffic lights at the intersection are controlled and planned based on the traffic congestion index to generate a preset control scheme. The actual traffic efficiency of the road segment is calculated based on the data from the vehicle terminal, and the actual traffic efficiency is compared with the expected efficiency of the preset control scheme to determine whether the preset control scheme is the optimal control scheme. If so, the traffic lights at the intersection are controlled according to the preset control scheme, and the optimal driving route is determined based on the data from the vehicle terminal. If not, or if there is an abnormal deviation between the traffic data and the vehicle terminal data, the control plan shall be determined based on previous data.

2. The intelligent traffic light control method based on multimodal data according to claim 1, characterized in that, The traffic data includes the number of vehicles, the number of times the vehicle travels at non-uniform speeds, and the total travel time of the vehicles; the vehicle terminal data includes the real-time cumulative fuel consumption of the vehicle passing through the monitored road segment, the actual passage time of the vehicle through the monitored road segment, and the length of the monitored road segment. The monitored road segment includes multiple adjacent intersections; The historical data includes the total number of vehicles, average traffic flow, average green light duration, and signal cycle for the historical time period corresponding to the target time period in the past week.

3. The intelligent traffic light control method based on multimodal data according to claim 2, characterized in that, Based on the traffic data, a traffic congestion index is calculated, and based on the traffic congestion index, traffic light control plans are developed at intersections to generate preset control schemes, including: The first data is calculated using the formula p1=n×t / m; where p1 is the first data representing the degree of road segment disturbance; n is the sum of the number of non-uniform speed travels of all vehicles in the monitored road segment; t is the cumulative travel time of all vehicles in the monitored road segment; and m is the number of vehicles in the monitored road segment. The first data is filtered using a preset first threshold range, and the filtered first data is obtained after removing outliers. Based on the filtered first data, the control parameters of the traffic lights at the intersection are adjusted to generate a preset control scheme. The adjustment operations of the control parameters include increasing the green light duration, decreasing the green light duration, adjusting the signal period, and adjusting the phase difference.

4. The intelligent traffic light control method based on multimodal data according to claim 2, characterized in that, The method for determining whether a preset control scheme is the optimal control scheme specifically includes: The second data is calculated using the formula p2=t' / s; where p2 is the second data representing the travel time per unit distance; t' is the actual travel time of a single vehicle through the monitored road segment; and s is the length of the monitored road segment. The second data is filtered using a preset second threshold range to obtain the filtered second data; Calculate the difference between the filtered second data and the theoretical expected travel time ratio corresponding to the preset control scheme. If the absolute value of the difference is within the preset error range, the preset control scheme is determined to be effective and is identified as the optimal control scheme.

5. The intelligent traffic light control method based on multimodal data according to claim 4, characterized in that, Determining the optimal driving route based on the data from the vehicle terminal specifically includes: The third data point is calculated using the formula p3 = l / s; Where p3 is the third data representing energy consumption per unit distance; l is the real-time cumulative fuel consumption of the vehicle passing through the monitored road segment; and s is the length of the monitored road segment. The third data is filtered using a preset third threshold range to obtain filtered third data; By combining the filtered second data and the filtered third data, different paths are evaluated using a weighted average to determine the optimal driving route.

6. The intelligent traffic light control method based on multimodal data according to claim 5, characterized in that, The method for obtaining the real-time cumulative fuel consumption value l of the vehicle passing through the monitored road section is as follows: The instantaneous fuel injection quantity or instantaneous energy consumption data of the target vehicle can be read in real time through the on-board diagnostic system or CAN bus interface installed on the target vehicle. By combining the vehicle's real-time GPS location information, a time window is extracted between the moment the vehicle enters the monitored road segment and the moment it leaves the monitored road segment; The instantaneous fuel injection quantity or instantaneous energy consumption data within the time window are integrated to obtain the real-time cumulative fuel consumption value of the vehicle passing through the monitored road segment.

7. A device / equipment / system for intelligent control of traffic lights based on multimodal data, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-6.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1-6.

Images