Street lamp control system and method for dynamic dimming in traffic flow direction

By integrating the traffic flow acquisition module and dimming module on the street lights, and adjusting the street light lighting using machine learning, the speeding problem caused by high brightness when there are fewer vehicles is solved, and safe traffic and energy consumption optimization is achieved.

CN118984514AInactive Publication Date: 2024-11-19HUNAN YIFU LIGHTING TECHNOLOGY CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411268099.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing street light control system has high brightness when there are fewer vehicles, causing the driver to speed up and increase the probability of traffic accidents.

Method used

The street light control system adopts dynamic dimming of traffic flow, and by loading traffic flow acquisition modules and dimming modules on smart street lights, using infrared sensors or high-definition cameras to collect vehicle, pedestrian and light intensity data, and generate dimming strategies based on machine learning models. , dynamically adjust the lighting intensity and direction of the street lights to ensure that the lighting intensity meets the requirements of the road surface, and reduce the lighting speed when speeding is over to remind the driver.

Benefits of technology

It effectively reduces the possibility of night curve accidents, improves traffic safety, optimizes energy consumption through dynamic dimming strategies, ensures that the lighting intensity of street lights meets the needs of vehicles and pedestrians, and reduces vehicle speeding behavior.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118984514A_ABST
    Figure CN118984514A_ABST
Patent Text Reader

Abstract

The invention discloses a traffic flow direction dynamic dimming street lamp control system and method, and relates to the technical field of street lamp control, and the system comprises an intelligent street lamp which is used for providing an illumination light source for a road surface, a traffic flow collection module is loaded on the intelligent street lamp, the traffic flow collection module is used for collecting road surface data, and a dimming module is loaded on the intelligent street lamp. The traffic flow acquisition module transmits road surface data to the dimming module, and the dimming module automatically generates a dimming strategy based on the road surface data; the dimming module controls the intelligent street lamp based on the dimming strategy and limits the speed of vehicles on the road surface at the same time; meanwhile, a driver of the vehicle is reminded to control the speed of the vehicle, the driver is supervised and urged to control the speed of the vehicle within a normal range, and safe traffic is facilitated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of street lamp control, and in particular to a street lamp control system and method for dynamic dimming of traffic flow direction. Background Art

[0002] Traditional street light control systems only control the brightness of street lights based on traffic flow. For example, patent publication number CN115884473A discloses an intelligent street light system and a smart street light including the system, and patent publication number CN114126157A discloses an intelligent street light monitoring system and method based on wireless sensor network adaptive dimming. When such street light control systems control street lights, if there are vehicles on the road surface corresponding to the street light, the street light control system will increase the brightness of the street light. If there are no vehicles on the road surface, the street light control system will reduce the brightness of the street light or turn off the street light. It is rare that there are no vehicles on the road surface at all. Therefore, the energy saving of street lights is very limited. In particular, when there are fewer vehicles, the brightness of the street lights is high, which causes drivers to speed and indirectly leads to the probability of traffic accidents. Summary of the invention

[0003] In order to overcome the above-mentioned technical problems, the purpose of the present invention is to provide a street lamp control system and method with dynamic dimming according to traffic flow, so as to solve the problem in the prior art that when there are fewer vehicles, the high brightness of the street lamps causes drivers to speed, resulting in a high probability of traffic accidents.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] Specifically, a street lamp control system with dynamic dimming according to traffic flow is provided, including a smart street lamp, which is used to provide lighting source for the road surface. A traffic flow collection module is installed on the smart street lamp, and the traffic flow collection module is used to collect road surface data. A dimming module is installed on the smart street lamp, and the traffic flow collection module transmits the road surface data to the dimming module, and the dimming module automatically generates a dimming strategy based on the road surface data; the dimming module controls the opening direction of the smart street lamp based on the dimming strategy, and the opening direction includes straight direction, right turn direction and left turn direction.

[0006] As a further solution of the present invention: the traffic flow collection module is an infrared sensor or a high-definition camera.

[0007] As a further solution of the present invention: the road surface data includes vehicle data, speed limit data, pedestrian data and road surface light intensity data.

[0008] As a further solution of the present invention: the dimming module automatically generates a dimming strategy based on road surface data in the following manner:

[0009] The dimming module collects historical training data sets in advance, and trains a machine learning model for predicting the lighting intensity of smart street lamps based on the historical training data sets;

[0010] The traffic flow collection module collects vehicle data, pedestrian data and road surface light intensity data in real time, and transmits the vehicle data, pedestrian data and road surface light intensity data to the dimming module. The dimming module uses a machine learning model to predict the opening direction of the smart street lamp based on the vehicle data, pedestrian data and road surface light intensity data.

[0011] As a further solution of the present invention: the historical training data set includes N groups of training data, N is a positive integer, each group of training data includes feature data and label data, wherein the feature data includes vehicle data, speed limit data, pedestrian data and road surface light intensity data;

[0012] The vehicle data is the vehicle information on the road surface collected by the traffic flow collection module when collecting each set of training data;

[0013] The speed limit data is the maximum vehicle speed allowed on the road surface when collecting each set of training data;

[0014] The pedestrian data is the pedestrian reflectivity of the road surface collected by the traffic flow collection module when collecting each set of training data;

[0015] The road surface illumination intensity data is the road surface illumination intensity value collected by the traffic flow collection module when collecting each set of training data;

[0016] The label data is the direction in which the smart street light is turned on when collecting each set of training data, where the straight direction is marked with a value of 0, the right turn direction is marked with a value of 1, and the left turn direction is marked with a value of -1.

[0017] As a further solution of the present invention: the vehicle information includes vehicle reflectivity, vehicle speed, vehicle moving direction and vehicle position.

[0018] As a further solution of the present invention: the dimming module numbers the smart street lamps, and the dimming module controls the turning-on direction and lighting intensity of the smart street lamps based on the numbers of the smart street lamps, and the lighting intensity forms a lighting speed based on the startup speed of the smart street lamps with corresponding numbers.

[0019] As a further solution of the present invention: the dimming module trains a machine learning model for predicting the lighting intensity of the smart street lamp in the following manner:

[0020] The feature data in each set of training data is converted into a feature vector, and the feature vector is used as the input of the machine learning model. The opening direction predicted by the machine learning model for the feature data is used as the output. The opening direction in the label data corresponding to each set of feature data is used as the prediction target. The training target is to minimize the sum of all prediction accuracies. The calculation formula for prediction accuracy is ui = (αi-βi) 2 , where ui is the prediction accuracy, αi is the predicted opening direction corresponding to the i-th group of feature data, and βi is the opening direction in the i-th group of label data; the machine learning model is trained until the prediction accuracy reaches convergence and the training is stopped.

[0021] As a further solution of the present invention: the dimming module presets a threshold of the lighting speed based on the speed limit data, and the threshold of the lighting speed is less than or equal to the speed limit data.

[0022] Specifically, a method for controlling a street lamp with dynamic dimming in the direction of traffic flow is provided, which is implemented based on the above-mentioned street lamp control system with dynamic dimming in the direction of traffic flow, and includes the following steps:

[0023] S1: The dimming module collects historical training data sets in advance;

[0024] S2: Based on the historical training data set, a machine learning model is trained to predict the light intensity of smart street lamps;

[0025] S3: The traffic flow collection module collects vehicle data, pedestrian data and road surface light intensity data in real time, and transmits the vehicle data, pedestrian data and road surface light intensity data to the dimming module;

[0026] S4: The dimming module uses a machine learning model to predict the direction in which the smart street light will be turned on based on vehicle data, pedestrian data, and road light intensity data;

[0027] S5: The dimming module numbers the smart street lamps. The dimming module controls the turning-on direction and lighting intensity of the smart street lamps based on the numbers of the smart street lamps. The lighting intensity forms a lighting speed based on the starting speed of the smart street lamps with corresponding numbers.

[0028] S6: The dimming module presets a threshold of the lighting speed based on the speed limit data, and limits the speed of the road vehicle based on the threshold of the lighting speed.

[0029] Beneficial effects of the present invention:

[0030] 1. In the present invention, the dimming module can automatically generate a dimming strategy based on the road surface data. Specifically, the dimming module can directly control the power of the lamp body on the smart street lamp through a microprocessor to achieve the dimming control of the smart street lamp. When the road surface data shows that there are vehicles or pedestrians under the smart street lamp, the dimming module can increase the lighting intensity of the smart street lamp to ensure that the smart street lamp can provide sufficient lighting intensity for vehicles or pedestrians. The lighting intensity can be adaptively adjusted according to the vehicles, pedestrians, vehicle speed, number of vehicles or number of pedestrians to ensure that the lighting intensity is consistent with the road surface data under the smart street lamp. When the road surface data shows that there are no vehicles or pedestrians under the smart street lamp, the dimming module can increase the lighting intensity of the smart street lamp, or directly turn off the smart street lamp to achieve the purpose of saving energy consumption of the smart street lamp.

[0031] 2. In the present invention, the dimming module can continuously collect the speed of the vehicle through the traffic flow collection module during the driving process of the vehicle, and continuously update the change speed of the smart street lamp through the speed of the vehicle. In addition, it should be noted that the dimming module needs to collect the speed limit data of the road surface corresponding to the smart street lamp. At this time, the dimming module needs to control the lighting speed threshold of the smart street lamp so that the lighting speed threshold of the smart street lamp is less than or equal to the speed limit data. In this way, during the driving process of the vehicle, if speeding occurs, the lighting speed of the smart street lamp will be lower than the driving speed of the vehicle, and the vehicle will exceed the lighting area of ​​the smart street lamp, so as to remind the driver of the corresponding vehicle to control the speed and urge the driver to control the speed of the vehicle within a normal range, which is conducive to safe traffic.

[0032] 3. In the present invention, by setting up a street light control system, the smart street lights in the street light control system will be consistent with the vehicle's driving speed and direction under the control of the dimming module, and the installation height of the smart street lights and the lighting intensity generated can provide a reference for drivers of vehicles on both sides of the curve, because the drivers of the vehicles can accurately judge the position and speed of the vehicles in the opposite lane through the changes of the smart street lights, thereby ensuring that the drivers of the vehicles can make predictions in advance and ensure that the drivers of the vehicles can more accurately control their vehicles to enter the curve and meet other vehicles, greatly reducing the possibility of accidents on the curve at night and contributing to safe traffic. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below in conjunction with the accompanying drawings:

[0034] Figure 1 It is a principle flow chart of the street lamp control system with dynamic dimming of traffic flow direction of the present invention;

[0035] Figure 2 It is a transmission diagram of the traffic flow collection module in the present invention;

[0036] Figure 3 It is a flow chart of the street lamp control method of traffic flow dynamic dimming of the present invention;

[0037] Figure 4 This is a start-up diagram of the smart street light of the present invention when the vehicle is traveling in the straight direction;

[0038] Figure 5 This is the startup diagram of the smart street light of the present invention when the vehicle is turning right;

[0039] Figure 6 This is the startup diagram of the smart street light when the vehicle of the present invention is turning left. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Example 1

[0042] like Figure 1 - Figure 2 As shown, the present invention discloses a street lamp control system with dynamic dimming of traffic flow, including a smart street lamp, which is used to provide a lighting source for the road surface. It should be noted that in the process of upgrading the traffic system, those skilled in the art can use the existing street lamps to upgrade the existing street lamps to smart street lamps, thereby reducing the cost of upgrading the street lamp control system. The smart street lamp is mainly composed of a lamp pole, a lamp body and a microprocessor arranged inside the lamp body. The microprocessor can directly control the power of the lamp body to achieve the purpose of controlling the dimming of the smart street lamp. A traffic flow collection module is installed on the smart street lamp. The traffic flow collection module is an infrared sensor or a high-definition camera. The infrared sensor can collect vehicles and pedestrians on the road surface. Specifically, the infrared sensor emits infrared light, and the infrared light will have a It has a certain infrared reflectivity, and the infrared sensor can judge the vehicles or pedestrians on the road according to the infrared reflectivity. The high-definition camera collects the vehicle pictures or pedestrian pictures on the road by taking pictures, and then uses the convolutional neural network model to identify the vehicle pictures or pedestrian pictures. The smart street lamp is equipped with a dimming module. It should be noted that the dimming module can be set in the microprocessor. The traffic flow collection module transmits the road data to the dimming module. The dimming module automatically generates a dimming strategy based on the road data; the dimming module controls the opening direction of the smart street lamp based on the dimming strategy. The opening direction includes the straight direction, the right turn direction and the left turn direction. It should be noted that when the vehicle needs to go straight at the intersection, the dimming module controls the smart street lamp to turn on the smart street lamp in the straight direction. Figure 4 As shown; when the vehicle needs to turn right at the intersection, the dimming module controls the smart street light to turn on the smart street light in the right turn direction, such as Figure 5 As shown; when the vehicle needs to turn left at the intersection, the dimming module controls the smart street light to turn on the smart street light in the left turn direction, such as Figure 6 As shown;

[0043] When the dimming module collects the road data transmitted by the traffic flow collection module, the dimming module can automatically generate a dimming strategy based on the road data. Specifically, the dimming module can directly control the power of the lamp body on the smart street lamp through the microprocessor to achieve the dimming control of the smart street lamp;

[0044] Dimming refers to adjusting the lighting intensity of smart street lights. The lighting intensity is set by technicians in this field according to the model and specifications of the smart street lights to any intensity level, ensuring that the smart street lights can provide different levels of lighting intensity. For example, when the road surface data shows that there are vehicles or pedestrians under the smart street lights, the dimming module can increase the lighting intensity of the smart street lights to ensure that the smart street lights can provide sufficient lighting intensity for vehicles or pedestrians. The lighting intensity can be adaptively adjusted according to the vehicles, pedestrians, vehicle speeds, number of vehicles or number of pedestrians to ensure that the lighting intensity is consistent with the road surface data under the smart street lights. When the road surface data shows that there are no vehicles or pedestrians under the smart street lights, the dimming module can increase the lighting intensity of the smart street lights, or directly turn off the smart street lights, so as to save the energy consumption of the smart street lights.

[0045] In actual operation, the smart street lights can be adaptively started according to the driving of the vehicle. For example, when there is a vehicle in the road data collected by one of the smart street lights through the traffic flow collection module, the dimming module can simultaneously obtain the vehicle speed and the driving direction of the vehicle through the traffic flow collection module, and then take the smart street light as the starting point, and the 3-5 smart street lights behind the smart street light along the driving direction of the vehicle simultaneously increase the lighting intensity, and then continuously change the driving direction of the vehicle based on the vehicle speed. For example, taking 5 smart street lights as an example, the opening direction of 5 consecutive smart street lights can be consistent with the vehicle speed direction, and continuously change in the driving direction of the vehicle. In this way, the smart street lights can be consistent with the vehicle during the driving process, so that the vehicle is always illuminated by 5 consecutive smart street lights, which can provide sufficient lighting intensity for the vehicle. The control of the smart street light is controlled by the dimming module. The dimming module can continuously collect the speed of the vehicle through the traffic flow collection module during the driving process of the vehicle, and continuously update the lighting speed of the smart street light through the speed of the vehicle. In addition, it should be noted that the dimming module needs to collect the speed limit data of the road surface corresponding to the smart street light. For example, the speed limit data of the road surface is that the vehicle speed must not exceed 50km / h. At this time, the dimming module needs to control the lighting speed threshold of the smart street light so that the lighting speed threshold of the smart street light is less than or equal to 50km / h. In this way, if the vehicle is speeding during driving, the lighting speed of the smart street light will be lower than the driving speed of the vehicle, and the vehicle will exceed the lighting area of ​​the smart street light, so as to remind the driver of the vehicle to control the speed and urge the driver to control the speed of the vehicle within a normal range, which is conducive to safe traffic.

[0046] When a vehicle is driving at night, the driver's field of vision is also limited due to the limitation of sight. Especially on a curved road, the driver often cannot find the vehicle driving in the opposite lane in time. By setting up the street light control system, the smart street lights in the street light control system will be consistent with the driving speed and driving direction of the vehicle under the control of the dimming module, and the installation height and lighting intensity of the smart street lights can provide a reference for the drivers of the vehicles on both sides of the curve, because the driver of the vehicle can accurately judge the position and speed of the vehicle in the opposite lane through the change of the smart street lights, so as to ensure that the driver of the vehicle can make a prediction in advance and control his vehicle to enter the curve and meet more accurately, which greatly reduces the possibility of accidents in curves at night and contributes to safe traffic.

[0047] When overtaking at night, the driver can also accurately judge the position and speed of the vehicle in the opposite lane through the changes in the smart street lights, ensuring that the driver can choose the right time to overtake;

[0048] It should also be noted that when there is too much vehicle traffic on the two-way lanes of the road, the smart street lights will produce long-distance lighting, that is, the smart street lights within the driver's sight are all turned on. The driver can know through the smart street lights that there is too much vehicle traffic on the road, so he should keep a safe distance and follow the car instead of overtaking. This provides an accurate reference for the driver to drive at night and ensures safe traffic.

[0049] Example 2

[0050] like Figure 1 and Figure 2 As shown, the road surface data includes vehicle data, speed limit data, pedestrian data and road surface light intensity data. The road surface light intensity data refers to the light intensity value of the road surface. It should be noted that the traffic flow collection module also includes a photosensor, which directly collects the light intensity value. The traffic flow collection module transmits the collected vehicle data, speed limit data, pedestrian data and road surface light intensity data to the dimming module.

[0051] The dimming module automatically generates a dimming strategy based on road surface data in the following way:

[0052] The dimming module collects historical training data sets in advance, and trains a machine learning model for predicting the direction in which the smart street lamp will turn on based on the historical training data sets;

[0053] The historical training data set includes N groups of training data, where N is a positive integer. It should be noted that the historical training data set is collected by technicians in the actual use of smart street lamps. Each group of training data includes feature data and label data, where the feature data includes vehicle data, speed limit data, pedestrian data, and road light intensity data;

[0054] When collecting each set of training data, the traffic flow collection module collects vehicle information on the road surface. The vehicle information includes vehicle reflectivity, vehicle speed, vehicle moving direction and vehicle position. It should be noted that when the traffic flow collection module is an infrared sensor, the infrared sensor will emit infrared rays to the road surface below the smart street lamp. When a vehicle passes the smart street lamp, the surface of the vehicle will reflect infrared rays. The infrared sensor determines whether there is a vehicle passing by the road surface below the smart street lamp by the received vehicle reflectivity. The infrared sensor also senses the movement of the vehicle and obtains the vehicle speed, vehicle moving direction and vehicle position through the movement of the vehicle. The vehicle position refers to the lane position of the vehicle when it reaches the intersection. The lane usually includes a straight lane, a right turn lane and a left turn lane. Since the speed of the vehicle is generally changing, the smart street lamp can continuously collect the vehicle speed through the infrared sensor to ensure the continuous update of the vehicle speed. The infrared sensor can convert the collected vehicle reflectivity, vehicle speed and vehicle moving direction data into electrical signals and transmit them to the dimming module;

[0055] The speed limit data is the maximum vehicle speed allowed on the road surface when collecting each set of training data. The road surface usually sets the maximum speed for vehicles. For example, the speed limit on highways is 120km / h, but highways are generally not equipped with smart street lights. The speed limit on urban elevated roads is generally 80km / h, and the speed limit on ordinary roads in cities is generally within the range of 40km / h-60km / h. Therefore, the speed limit data has different values ​​based on different road surfaces and is specifically set according to local traffic regulations;

[0056] When collecting pedestrian data for each set of training data, the traffic flow collection module collects the pedestrian reflectivity of the road surface. Sidewalks are usually set on both sides of the road surface. When the infrared sensor on the traffic flow collection module collects the pedestrian reflectivity of the road surface, it means that there are pedestrians walking on the sidewalk. At this time, the traffic flow collection module converts the signal into an electrical signal and transmits it to the dimming module;

[0057] When collecting each set of training data, the traffic flow collection module collects the road surface light intensity value. It should be noted that the traffic flow collection module also includes a photosensor. The dimming module presets the threshold of the photosensor. When the light intensity value collected by the photosensor is greater than or equal to the threshold, it means that the road surface lighting source under the smart street lamp is sufficient. For example, during the day, the smart street lamp does not need to be turned on. If the light intensity value collected by the photosensor is less than the threshold, it means that the road surface lighting source under the smart street lamp is insufficient. At this time, the smart street lamp is adjusted according to the dimming strategy of the dimming module.

[0058] The label data is the direction in which the smart street lights are turned on when collecting each set of training data. It should be noted that the turning on directions include straight ahead, right turn, and left turn. When a vehicle needs to go straight at an intersection, the dimming module controls the smart street lights to turn on the smart street lights in the straight ahead direction. Figure 4 As shown; when the vehicle needs to turn right at the intersection, the dimming module controls the smart street light to turn on the smart street light in the right turn direction, such as Figure 5 As shown; when the vehicle needs to turn left at the intersection, the dimming module controls the smart street light to turn on the smart street light in the left turn direction, such as Figure 6 As shown, the straight direction is marked with a value of 0, the right turn direction is marked with a value of 1, and the left turn direction is marked with a value of -1.

[0059] The dimming module numbers the smart street lamps. For example, if there are 200 smart street lamps on Road A, the smart street lamps on Road A may be numbered A001, A002, ..., A200. Each number represents a smart street lamp. The dimming module controls the lighting intensity of the smart street lamp based on the number of the smart street lamp. The lighting intensity forms the lighting speed based on the startup speed of the smart street lamp with the corresponding number.

[0060] It should be noted that the dimming module can simultaneously obtain the vehicle speed and the vehicle's driving direction through the traffic flow collection module, and then take the smart street light as the starting point, and the 3-5 smart street lights behind the smart street light along the vehicle's driving direction simultaneously increase the lighting intensity, and then continuously change the driving direction of the vehicle based on the vehicle speed. For example, taking 5 smart street lights as an example (the data of the smart street lights can be determined based on the installation spacing between the smart street lights. When the spacing between the smart street lights is large, the number of smart street lights started at the same time is reduced, and vice versa), the direction of vehicle travel is the direction in which the smart street light number increases. When the traffic flow collection module on number A015 recognizes the vehicle, it is coded A015 is the starting point, and the following A016, A017, A018, and A019 will start at the same time. As the vehicle moves, the vehicle will come under A019 according to the speed of the vehicle. A019 will collect the speed of the vehicle again. In this process, the smart street lights after A019 will be started one by one according to the speed of the vehicle, and the smart street lights between A015-A019 will be turned off one by one according to the speed of the vehicle. This shows that when the vehicle is driving at a normal speed, the smart street lights can always be turned on on the route of the vehicle, ensuring that the smart street lights can match the driving of the vehicle and can be consistent with the vehicle, so that the vehicle is always illuminated by 5 consecutive smart street lights;

[0061] like Figure 5 As shown in the figure, when the speed of the vehicle exceeds the maximum speed limit of Road A, the startup speed of the smart street light is based on the maximum speed limit of Road A. Once the vehicle exceeds the range of the 5 consecutive smart street lights, that is, when the 5 consecutive smart street lights cannot detect the vehicle, the 5 consecutive smart street lights will be turned off at the same time, and the other smart street lights will detect the vehicle again and light up again. If the driver has been speeding, he will feel that the street lights are on and off, so as to warn the driver of the vehicle that he is speeding. As long as the vehicle is within the normal speed, the vehicle will feel that the street lights are always on during driving, and there will be a good driving feeling, so as to cultivate the driver's good driving habit of not exceeding the driving time.

[0062] It should also be noted that when the same smart street light receives two or more different commands from the dimming module, unless the control information received by the smart street light from the dimming module is all turned off, the smart street light needs to be turned on, that is, as long as the smart street light receives a control information to turn on, the smart street light needs to be turned on.

[0063] The dimming module trains a machine learning model to predict the direction in which the smart street light turns on as follows:

[0064] The feature data in each set of training data is converted into a feature vector, and the feature vector is used as the input of the machine learning model. The opening direction predicted by the machine learning model for the feature data is used as the output. The opening direction in the label data corresponding to each set of feature data is used as the prediction target. The training target is to minimize the sum of all prediction accuracies. The calculation formula for prediction accuracy is ui = (αi-βi) 2 , where ui is the prediction accuracy, αi is the predicted opening direction corresponding to the i-th group of feature data, and βi is the opening direction in the i-th group of label data; the machine learning model is trained until the prediction accuracy reaches convergence and the training is stopped. The converged label is set by a technician in this field according to the specific model training situation;

[0065] It should be noted that if the opening direction in the label data is straight ahead, the corresponding value is 0, and the predicted opening direction is a right turn. The prediction accuracy obtained according to the formula is 1, but we need the prediction accuracy to be 0. In this way, the prediction accuracy will not meet expectations, and the model needs to be trained until the predicted lighting intensity level is consistent with the lighting intensity level in the label data. That is, when ui=0, training is stopped.

[0066] The traffic flow collection module collects vehicle data, pedestrian data and road light intensity data in real time, and transmits the vehicle data, pedestrian data and road light intensity data to the dimming module. The dimming module uses a machine learning model based on the vehicle data, pedestrian data and road light intensity data to predict the lighting intensity of the smart street lamp. Then the dimming module controls the corresponding smart street lamp based on the lighting intensity to provide lighting for the road surface under the smart street lamp, ensuring that the smart street lamp can best match the vehicle and pedestrian conditions on the road, that is, the smart street lamp is turned on when there are cars and people, and turned off when there are no people and no cars, so as to achieve the effect of minimum energy consumption.

[0067] The dimming module presets the threshold of the lighting speed based on the speed limit data. The threshold of the lighting speed should be less than or equal to the speed limit data. The road surface usually sets the maximum speed for vehicles. For example, the speed limit on the highway is 120km / h, but smart street lights are generally not installed on highways. The speed limit on urban elevated roads is generally 80km / h, and the speed limit on ordinary roads in the city is generally within the range of 40km / h-60km / h. Therefore, the speed limit data has different values ​​based on different road surfaces, and is specifically set according to local traffic regulations. For example, when the speed limit data on Road A is 50km / h, the threshold of the lighting speed preset by the dimming module is 50km / h, and the speed of the smart street light change needs to be less than or equal to 50km / h.

[0068] Example 3

[0069] like Figure 1 - Figure 5 As shown, the present invention discloses a street lamp control method for dynamic dimming of traffic flow direction, comprising the following steps:

[0070] S1: The dimming module collects a set of historical training data in advance. The set of historical training data is collected by technicians in the actual use of smart street lamps. Each set of training data includes feature data and label data, where the feature data includes vehicle data, speed limit data, pedestrian data and road light intensity data;

[0071] Vehicle data refers to the vehicle information collected by the traffic flow collection module on the road when collecting each set of training data. The vehicle information includes vehicle reflectivity, vehicle speed and vehicle movement direction. The speed limit data refers to the maximum vehicle speed allowed on the road when collecting each set of training data. Pedestrian data refers to the pedestrian reflectivity collected by the traffic flow collection module when collecting each set of training data. Road illumination intensity data refers to the illumination intensity value of the road when collecting each set of training data. Label data refers to the direction in which the smart street lights are turned on when collecting each set of training data.

[0072] S2: Based on the historical training data set, a machine learning model for predicting the direction in which the smart street light is turned on is trained. The dimming module trains a machine learning model for predicting the direction in which the smart street light is turned on as follows:

[0073] The feature data in each set of training data is converted into a feature vector, and the feature vector is used as the input of the machine learning model. The opening direction predicted by the machine learning model for the feature data is used as the output. The opening direction in the label data corresponding to each set of feature data is used as the prediction target. The training target is to minimize the sum of all prediction accuracies. The calculation formula for prediction accuracy is ui = (αi-βi) 2 , where ui is the prediction accuracy, αi is the predicted opening direction corresponding to the i-th group of feature data, and βi is the opening direction in the i-th group of label data; the machine learning model is trained until the prediction accuracy reaches convergence and the training is stopped. The converged label is set by a technician in this field according to the specific model training situation;

[0074] It should be noted that if the opening direction in the label data is the straight direction, the corresponding value is 0, and the predicted opening direction is the right turn direction, then the prediction accuracy obtained according to the formula is 1, and we need the prediction accuracy to be 0, so the prediction accuracy will not meet expectations, and the model needs to be trained until the predicted lighting intensity level is consistent with the lighting intensity level in the label data, that is, when ui=0, the training is stopped;

[0075] S3: The traffic flow collection module collects vehicle data, pedestrian data and road surface light intensity data in real time, and transmits the vehicle data, pedestrian data and road surface light intensity data to the dimming module;

[0076] It should be noted that when the traffic flow collection module is an infrared sensor, the infrared sensor will emit infrared rays to the road surface under the smart street lamp. When a vehicle passes by the smart street lamp, the surface of the vehicle will reflect infrared rays. The infrared sensor determines whether there is a vehicle passing by the road surface under the smart street lamp by the received vehicle reflectivity. The infrared sensor also senses the movement of the vehicle, and obtains the vehicle speed and direction of movement through the movement of the vehicle. Since the speed of the vehicle is generally changing, the smart street lamp can continuously collect the vehicle speed through the infrared sensor to ensure the continuous update of the vehicle speed. The infrared sensor can convert the collected vehicle reflectivity, vehicle speed and vehicle direction and other data into electrical signals and transmit them to the dimming module; the road surface usually sets the maximum speed for vehicles. For example, the speed limit on highways is 120km / h, but smart street lamps are generally not installed on highways, and the speed limit on urban elevated roads is generally 80km / h. h. The speed limit on ordinary roads in cities is generally within the range of 40km / h-60km / h. Therefore, the speed limit data has different values ​​based on different road surfaces, and is specifically set according to local traffic regulations; sidewalks are usually set on both sides of the road surface. When the infrared sensor on the traffic flow collection module collects the pedestrian reflectivity of the road surface, that is, pedestrians appear on the sidewalk, the traffic flow collection module converts the signal into an electrical signal and transmits it to the dimming module; the traffic flow collection module also includes a photosensor, and the dimming module presets the threshold of the photosensor. When the light intensity value collected by the photosensor is greater than or equal to the threshold, it means that the road lighting source under the smart street lamp is sufficient. For example, during the day, the smart street lamp does not need to be turned on. If the light intensity value collected by the photosensor is less than the threshold, it means that the road lighting source under the smart street lamp is insufficient. At this time, the smart street lamp is adjusted according to the dimming strategy of the dimming module;

[0077] The lighting intensity is set to any level according to the model and specifications of the smart street lamp by technicians in this field, so as to ensure that the smart street lamp can provide different levels of lighting intensity. For example, the lighting intensity of the smart street lamp can be divided into three levels. The first level is when there are no pedestrians or vehicles under the smart street lamp, the lighting intensity of the smart street lamp is low. The low level can refer to the smart street lamp being turned off, that is, the smart street lamp does not generate lighting intensity, saving the energy consumption of the smart street lamp. The second level is when there are only pedestrians or vehicles under the smart street lamp, the lighting intensity of the smart street lamp is medium. The medium level can refer to the smart street lamp turning on the lamp body with medium power, that is, half of the rated power of the lamp body. The technicians in this field can also make an adaptive selection according to the specifications and power of the lamp body to provide a medium level of lighting intensity for the road surface under the smart street lamp. The third level is when there are both pedestrians and vehicles under the smart street lamp, the lighting intensity of the smart street lamp is high. The high level can refer to the smart street lamp turning on the lamp body with high power, that is, the rated power of the lamp body. The technicians in this field can also make an adaptive selection according to the specifications and power of the lamp body to provide a high level of lighting intensity for the road surface under the smart street lamp.

[0078] S4: The dimming module uses a machine learning model to predict the direction in which the smart street light will be turned on based on vehicle data, pedestrian data, and road light intensity data;

[0079] S5: The dimming module numbers the smart street lamps. The dimming module controls the opening direction and lighting intensity of the smart street lamps based on the numbers of the smart street lamps. The lighting intensity forms a lighting speed based on the starting speed of the smart street lamps with corresponding numbers. For example, if there are 200 smart street lamps on Road A, the smart street lamps on Road A can be numbered A001, A002, ..., A200, and each number represents a smart street lamp.

[0080] Taking 5 smart street lights as an example, the direction of vehicle travel is the direction of increasing smart street light numbers. When the traffic flow collection module on number A001 recognizes the vehicle, it will take number A001 as the starting point, and subsequent numbers A002, A003, A004, and A005 will be started at the same time. As the vehicle travels, the vehicle will come under number A005 according to the driving speed, and number A005 will collect the vehicle speed again. In this process, the smart street lights after number A005 are started one by one at the speed of the vehicle, and the smart street lights between A001-A005 are turned off one by one at the speed of the vehicle, ensuring that the smart street lights can match the driving of the vehicle and can be consistent with the vehicle, so that the vehicle is always illuminated by 5 consecutive smart street lights.

[0081] S6: The dimming module presets a threshold of the lighting speed based on the speed limit data, and limits the speed of vehicles on the road based on the threshold of the lighting speed. For example, when the speed limit data of Road A is 50km / h, the threshold of the lighting speed preset by the dimming module is 50km / h. The speed of the smart street light change needs to be less than or equal to 50km / h. If the vehicle speed is greater than 50km / h, and the smart street light still changes at a speed of 50km / h, the vehicle will eventually exceed the lighting area. When 5 consecutive smart street lights cannot collect the vehicle, the 5 consecutive smart street lights will be turned off at the same time, and the other smart street lights will collect the vehicle again and light up again. If the driver has been speeding, he will feel that the street lights are on and off, so as to warn the driver of the vehicle that he is speeding. As long as the vehicle is at a normal speed, the vehicle will feel that the street lights are always on during driving, and there will be a good driving feeling, so as to cultivate the driver's good driving habit of not exceeding the time limit.

[0082] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A street light control system with dynamic dimming according to traffic flow, characterized in that: include: Smart street lights, which are used to provide lighting sources for the road surface; Traffic flow collection module, which is installed on the smart street light and is used to collect road surface data; The dimming module is installed on the smart street light. The traffic flow collection module transmits the road surface data to the dimming module, and the dimming module automatically generates a dimming strategy based on the road surface data. The dimming module controls the turning-on direction of the smart street lamp based on the dimming strategy, and the turning-on direction includes a straight direction, a right turn direction, and a left turn direction.

2. The street light control system with dynamic dimming according to traffic flow direction according to claim 1, characterized in that: The traffic flow collection module is an infrared sensor or a high-definition camera.

3. The street lamp control system with dynamic dimming according to traffic flow according to claim 1, characterized in that: The road surface data includes vehicle data, speed limit data, pedestrian data and road surface light intensity data.

4. The street lamp control system with dynamic dimming according to traffic flow direction according to claim 3 is characterized in that: The dimming module automatically generates a dimming strategy based on road surface data in the following manner: The dimming module collects historical training data sets in advance, and trains a machine learning model for predicting the lighting intensity of smart street lamps based on the historical training data sets; The traffic flow collection module collects vehicle data, pedestrian data and road surface light intensity data in real time, and transmits the vehicle data, pedestrian data and road surface light intensity data to the dimming module. The dimming module uses a machine learning model to predict the opening direction of the smart street lamp based on the vehicle data, pedestrian data and road surface light intensity data.

5. The street lamp control system with dynamic dimming according to traffic flow direction according to claim 4, characterized in that: The historical training data set includes N groups of training data, where N is a positive integer, and each group of training data includes feature data and label data, wherein the feature data includes vehicle data, speed limit data, pedestrian data, and road surface light intensity data; The vehicle data is the vehicle information on the road surface collected by the traffic flow collection module when collecting each set of training data; The speed limit data is the maximum vehicle speed allowed on the road surface when collecting each set of training data; The pedestrian data is the pedestrian reflectivity of the road surface collected by the traffic flow collection module when collecting each set of training data; The road surface illumination intensity data is the road surface illumination intensity value collected by the traffic flow collection module when collecting each set of training data; The label data is the direction in which the smart street light is turned on when collecting each set of training data, where the straight direction is marked with a value of 0, the right turn direction is marked with a value of 1, and the left turn direction is marked with a value of -1.

6. The street light control system with dynamic dimming according to traffic flow direction according to claim 5, characterized in that: The vehicle information includes vehicle reflectivity, vehicle speed, vehicle moving direction and vehicle position.

7. The street light control system with dynamic dimming according to traffic flow direction according to claim 6, characterized in that: The dimming module numbers the smart street lamps, and controls the turning-on direction and lighting intensity of the smart street lamps based on the numbers of the smart street lamps. The lighting intensity forms a lighting speed based on the starting speed of the smart street lamps with corresponding numbers.

8. The street lamp control system with dynamic dimming according to traffic flow according to claim 5, characterized in that: The dimming module trains a machine learning model for predicting the direction in which the smart street lamp is turned on as follows: The feature data in each set of training data is converted into a feature vector, and the feature vector is used as the input of the machine learning model. The opening direction predicted by the machine learning model for the feature data is used as the output. The opening direction in the label data corresponding to each set of feature data is used as the prediction target. The training target is to minimize the sum of all prediction accuracies. The calculation formula for prediction accuracy is ui = (αi-βi) 2 , where ui is the prediction accuracy, αi is the predicted opening direction corresponding to the i-th group of feature data, and βi is the opening direction in the i-th group of label data; the machine learning model is trained until the prediction accuracy reaches convergence and the training is stopped.

9. The street lamp control system with dynamic dimming according to traffic flow according to claim 7, characterized in that: The dimming module presets a threshold of the lighting speed based on the speed limit data, and the threshold of the lighting speed is less than or equal to the speed limit data.

10. A method for controlling a street lamp with dynamic dimming according to traffic flow direction, which is implemented based on the street lamp control system with dynamic dimming according to traffic flow direction according to claim 9, characterized in that: The following steps are involved: S1: The dimming module collects historical training data sets in advance; S2: Based on the historical training data set, a machine learning model is trained to predict the light intensity of smart street lamps; S3: The traffic flow collection module collects vehicle data, pedestrian data and road surface light intensity data in real time, and transmits the vehicle data, pedestrian data and road surface light intensity data to the dimming module; S4: The dimming module uses a machine learning model to predict the direction in which the smart street light will be turned on based on vehicle data, pedestrian data, and road light intensity data; S5: The dimming module numbers the smart street lamps. The dimming module controls the turning-on direction and lighting intensity of the smart street lamps based on the numbers of the smart street lamps. The lighting intensity forms a lighting speed based on the starting speed of the smart street lamps with corresponding numbers. S6: The dimming module presets a threshold of the lighting speed based on the speed limit data, and limits the speed of the road vehicle based on the threshold of the lighting speed.

Citation Information

Patent Citations

  • Streetlight network control method and network structure thereof

    CN102421219A

  • Running speed reminding method based on street lamp radio-frequency technology

    CN106228814A

  • Energy-saving street lamp capable of interconnecting and vibrating automatic switch with thermal induction through active infrared intrusion communication

    CN110822344A

  • Smart city street lamp intelligent control method and system based on Internet of Things

    CN115297595A

  • Intelligent lighting control method and system based on real-time environment information

    CN115767842A