CN115662152A - Urban traffic management self-adaptive system based on deep learning drive - Google Patents
Urban traffic management self-adaptive system based on deep learning drive Download PDFInfo
- Publication number
- CN115662152A CN115662152A CN202211178762.1A CN202211178762A CN115662152A CN 115662152 A CN115662152 A CN 115662152A CN 202211178762 A CN202211178762 A CN 202211178762A CN 115662152 A CN115662152 A CN 115662152A
- Authority
- CN
- China
- Prior art keywords
- preset
- current intersection
- intersection
- period
- control module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to an urban traffic management self-adaptive system based on deep learning drive, which comprises a model construction module and a model construction module, wherein the model construction module comprises a storage unit and a model construction unit, and the storage unit is used for storing the vehicle flow of each road junction in each time period; the acquisition module comprises a plurality of acquisition units, is arranged at each intersection and is used for acquiring images at each intersection; the preprocessing module is connected with the acquisition module, acquires information of vehicles to be passed in an image preset area of the current intersection, acquires the urgency of the current intersection, compares the acquired urgency of the current intersection with the preset urgency, and predicts the blockage level of the current intersection; and the control module is connected with the model building model, the acquisition module and the preprocessing module and is used for adjusting the green signal ratio in the current intersection traffic light switching period and the current intersection traffic light switching period according to the current intersection jam level.
Description
Technical Field
The invention relates to the field of urban traffic management, in particular to an urban traffic management self-adaptive system based on deep learning drive.
Background
Traffic congestion is a major problem in many cities. The large number of vehicles on the road that are driving each day makes the task of managing traffic more cumbersome. Extensive research is being conducted to make traffic management systems more adaptive and intelligent.
And installing cameras on roads and intersections, and implementing autonomous punishment and personal identification on violation of traffic rules. In terms of traffic management, most traffic intersections use a fixed-time green light cycle system to manage traffic. However, the conventional operation of adding a fixed cycle period to the traffic light management system has certain limitations, and is very inefficient in adjusting traffic congestion, and the conventional system lacks intelligent management, so that the driver is required to wait for the traffic light to turn green to drive even when there is no vehicle at another intersection, and the inevitable waiting time is uneasy, often violates traffic regulations and causes accidents.
In order to manage traffic congestion and to implement an automated traffic management method, researchers have been working on a sense network function to accumulate traffic congestion and a range of vehicles traveling at each intersection to make a decision when estimating a length of time that a traffic light can be kept green, but this method has a disadvantage that it is still more manual and time-consuming. In the above-mentioned research, the traffic flow video is generally processed as a standard input data. However, video is a collection of images, so all images require real-time processing, which is an expensive and time-consuming calculation, in order to process the video and gain insight into the flow. Furthermore, the continuous processing of images extracted from the video can affect the persistence of the system. Because of these limitations, there is a great need for a fast, accurate and cost-effective intelligent traffic management.
Chinese patent ZL202110968627.6 discloses a traffic monitoring system and method based on intelligent traffic internet of things, which automatically complete the adjustment of signal lamp time of each lane by collecting real-time information of motor vehicles, non-motor vehicles and pedestrians, thereby improving crossing traffic efficiency, but still does not solve the technical problem of crossing vehicle congestion.
Disclosure of Invention
Therefore, the urban traffic management self-adaptive system based on deep learning drive can solve the problem that the congestion condition cannot be acquired according to the current intersection urgency degree so as to adjust the green-to-traffic ratio of the intersection, so that the current intersection traffic management meets the preset standard.
In order to achieve the above object, the present invention provides an adaptive system for urban traffic management based on deep learning driving, comprising:
the model building module comprises a storage unit and a model building unit, wherein the storage unit is used for storing vehicle flow of each road junction in each time period, the model building unit is used for building a traffic signal lamp conversion period model according to the vehicle flow of each road junction stored in the storage unit, and the model building unit regulates the traffic signal lamp conversion period according to the green light utilization rate;
the acquisition module comprises a plurality of acquisition units, is arranged at each intersection and is used for acquiring images at each intersection;
the preprocessing module is connected with the acquisition module and used for preprocessing the images of the intersections acquired by the acquisition module and predicting the blockage levels of the intersections, wherein the preprocessing module acquires the information of vehicles to be passed in the preset image area of the current intersection to acquire the urgency of the current intersection and compares the acquired urgency of the current intersection with the preset urgency to predict the blockage levels of the current intersection;
and the control module is connected with the model building model, the acquisition module and the preprocessing module and is used for adjusting the green signal ratio in the current intersection traffic light conversion period and the current intersection traffic light conversion period according to the current intersection jam level, wherein if the current intersection jam level is A1 level, the control module reduces the current intersection traffic light conversion period, if the current intersection jam level is A3 level, the control module improves the green signal ratio in the current intersection traffic light conversion period, and meanwhile, the control module compares the obtained current intersection green signal ratio with a preset green signal ratio standard value and adjusts the green signal ratio in the traffic light conversion period of the next intersection and the previous intersection so as to enable the traffic condition of the current intersection to accord with a preset standard.
Furthermore, the model construction unit presets a flow L, and selects each time period cycle to construct a current intersection traffic light switching cycle module according to the comparison between the acquired vehicle flow lk and the preset flow in each time period of the current intersection, wherein,
when lk is less than or equal to L1, the model construction unit selects a first preset period T1 as a kth time period traffic light conversion period;
when L1 is larger than lk and is smaller than or equal to L2, the model construction unit selects a second preset period T2 as a kth time period traffic light conversion period;
when the Lk is more than L2 and less than or equal to L3, the model construction unit selects a third preset period T3 as a kth time period traffic light conversion period;
when lk is larger than L3, the model construction unit selects a fourth preset period T4 as a k-th time period traffic signal lamp conversion period;
the flow L is preset by the model building unit, a first preset flow L1, a second preset flow L2, a third preset flow L3 and a fourth preset flow L4 are set, the period T is preset by the model building unit, and a first preset period T1, a second preset period T2, a third preset period T3 and a fourth preset period T4 are set, k =1,2,. N and n are the total time period.
Further, the preprocessing module acquires the current intersection urgency degree p according to the distance s between the last vehicle in the preset area and the first vehicle at the intersection in the image acquired by the acquisition module and the vehicle running average speed v0, sets p = hj x (s/v 0)/Tig, compares the acquired current intersection urgency degree with the preset urgency degree, and predicts the current intersection blockage grade, wherein,
when P is less than or equal to P1, the preprocessing module predicts that the current intersection blockage level is a first preset blockage level A1;
when P1 is larger than P and smaller than P2, the preprocessing module predicts that the current intersection blockage grade is a second preset blockage grade A2;
when P is larger than or equal to P2, the preprocessing module predicts that the current intersection blockage grade is a third preset blockage grade A3;
the pre-processing module is used for presetting a congestion grade A, setting a first preset congestion grade A1, a second preset congestion grade A2 and a third preset congestion grade A3, hj is an urgency degree compensation parameter, and i =1,2,3,4 and Tig is the green time in the conversion period of the Ti traffic signal lamp.
Furthermore, the preprocessing module presets a vehicle density M, acquires the vehicle density M in the current preset area, compares the vehicle density M with the preset vehicle density, selects a urgency degree compensation parameter, wherein,
when M is less than or equal to M1, the preprocessing module selects a first preset urgency degree compensation parameter h1 as a current intersection urgency degree compensation parameter;
when M1 is larger than M and smaller than M2, the preprocessing module selects a second preset urgency degree compensation parameter h2 as a current intersection urgency degree compensation parameter;
when M is larger than or equal to M2, the preprocessing module selects a third preset urgency degree compensation parameter h3 as a current intersection urgency degree compensation parameter;
the vehicle density M is preset by the preprocessing module, a first preset vehicle density M1, a second preset vehicle density M2, a preset tightness compensation parameter h, a first preset tightness compensation parameter h1, a second preset tightness compensation parameter h2 and a third preset tightness compensation parameter h3 are set.
Further, the control module adjusts the traffic light switching period and the green signal ratio in the period of the current intersection according to the congestion level of the current intersection predicted by the preprocessing module, wherein,
if the current intersection congestion level is A1, the control module reduces the current intersection traffic light switching period Ti to Ti1, and sets Ti1= Ti x (1- (P1-P)/P1);
if the current intersection congestion level is A2, the control module does not adjust the current intersection traffic light switching period and the green signal ratio in the period;
if the current intersection congestion level is A3, the control module increases the split green ratio Tig to Tig1 in the current intersection traffic light conversion period, and sets Tig1= Tig x (1 + (P-P2)/P2).
Further, the control module obtains the green signal ratio Tig1 in the traffic light switching period of the current intersection and compares the green signal ratio Tig1 with a preset green signal ratio standard value T0 of the control module, and adjusts the green signal ratio in the traffic light switching period of the previous intersection and the next intersection, wherein,
when the Tig1 is less than or equal to T0, the control module does not adjust the split ratio of the traffic signal lamps at other intersections in the conversion period;
when the Tig1 is larger than T0, the control module improves the split ratio in the traffic signal lamp conversion period of the next intersection and shortens the split ratio in the traffic signal lamp conversion period of the previous intersection.
Further, the control module obtains that the split green ratio in the traffic light conversion period of the current intersection is larger than a preset split green ratio standard value, the control module increases the split green ratio Tx to Tx1 in the traffic light conversion period of the next intersection, sets Tx1= Tx x (1 + (Tig 1-T0)/T0), shortens the split green ratio Ts to Ts1 in the traffic light conversion period of the previous intersection, and sets Ts1= Ts x (1- (Tig 1-T0)/T0).
Further, the control module compares the acquired green light utilization rate F of the current intersection with a preset green light utilization rate F0, and adjusts a preset area of the preprocessing module and a preset traffic signal light conversion period of the model construction unit, wherein,
when F is less than or equal to F1, the control module reduces a preset traffic signal lamp conversion period Ti to Ti2 of the model building unit, and sets Ti2= Ti x (1- (F1-F)/F1);
when the F1 is larger than the F and smaller than the F1, the control module does not adjust the preset area;
when F is larger than or equal to F2, the control module enlarges a preset area of the preprocessing module;
the control module presets a green light utilization rate F, sets a first preset green light utilization rate F1, and presets a second green light utilization rate F2.
Further, the control module obtains a green light utilization rate f of the current intersection and sets f = d/Ti, wherein d is the number of vehicles passing through the current intersection in the current traffic light conversion period.
Further, when the control module obtains that the utilization rate of the green light at the current intersection is greater than or equal to a second preset utilization rate of the green light, the control module enlarges a preset area w to w1 of the preprocessing module, and sets w1= w x (1 + (F-F2)/F2).
Compared with the prior art, the traffic signal lamp conversion cycle adjusting method has the beneficial effects that the control module is arranged and used for adjusting the split green ratio in the current intersection traffic signal lamp conversion cycle and the current intersection traffic signal lamp conversion cycle according to the current intersection jam level, wherein if the current intersection jam level is A1 level, the control module reduces the current intersection traffic signal lamp conversion cycle, if the current intersection jam level is A3 level, the control module improves the split green ratio in the current intersection traffic signal lamp conversion cycle, meanwhile, the control module compares the obtained split green ratio with the preset split green ratio standard value and adjusts the split green ratio in the next intersection traffic signal lamp conversion cycle and the last intersection traffic signal lamp conversion cycle, so that the traffic condition of the current intersection meets the preset standard.
Particularly, the model construction unit divides the preset period into four clear standards, the model construction unit compares the current intersection vehicle flow in each time period stored in the storage unit with the preset flow, and selects the optimal flow as the current intersection traffic light conversion period in each time period, wherein if the vehicle flow in the current time period of the current intersection acquired by the model construction unit is less than or equal to the first preset flow, the model construction module selects a smaller period as the current intersection traffic light conversion period in the current time period, and so on, the model construction module compares the acquired flow with the preset flow, and selects a corresponding period as the current time period traffic light conversion period, and the model construction unit constructs the traffic light conversion period model according to the acquired traffic light conversion period in each time period of the current intersection.
Particularly, the invention obtains the current intersection urgency level according to the total length of vehicles in a preset area and the average speed of vehicle running by arranging a preprocessing module, and obtains the congestion level of the current intersection according to the current intersection urgency level and the preset urgency level by comparing the current intersection urgency level with the preset vehicle density, wherein the preprocessing module selects an optimal urgency level compensation parameter to compensate the urgency level according to the comparison between the vehicle density in the preset area and the preset vehicle density, so as to avoid inaccurate urgency level obtaining caused by short time of running at the average speed of vehicle running due to short vehicle gaps caused by too low vehicle density, when the preprocessing module obtains the current intersection urgency level which is less than or equal to the first preset urgency level, the current intersection is indicated as a small number of vehicles, the preprocessing module predicts the current intersection congestion level as a level 1, when the preprocessing module obtains the current intersection urgency level which is between the first preset urgency level and the second preset urgency level, the current intersection is indicated as a vehicle intersection, more preprocessing modules predict the current congestion level as a level 2, and when the preprocessing module obtains the current intersection urgency level which is greater than or equal to the second preset urgency level, the current intersection is indicated as a level, and the current intersection is indicated as a current intersection congestion level which is indicated as a congestion level 3.
Particularly, the green signal ratio in the conversion period of the traffic signal lamp at the current intersection is adjusted according to the grade of the congestion of the current intersection predicted by the preprocessing module, wherein when the grade of the congestion of the current intersection predicted by the preprocessing module is A1 grade, the fact that the congestion degree of the current intersection is low is indicated, in order to improve the efficiency of traffic conversion, the control module reduces the conversion period of the traffic signal lamp at the current intersection, when the grade of the congestion of the current intersection predicted by the preprocessing module is A2, the fact that the congestion condition of the current intersection is in a standard range is indicated, the control module does not adjust the green signal ratio in the conversion period and the conversion period of the traffic signal lamp at the current intersection, when the grade of the congestion of the current intersection predicted by the preprocessing module is A3, the fact that the current intersection is about to be seriously congested is indicated, and the control module improves the green signal ratio in the conversion period of the traffic signal lamp at the current intersection so as to solve the possible congestion problem at the current intersection.
Particularly, the control module is provided with a green signal ratio standard value, and compares the obtained green signal ratio in the traffic signal conversion period of the current intersection with a preset green signal ratio standard value to adjust the green signal ratio of other intersections, wherein if the green signal ratio in the traffic signal conversion period of the current intersection is less than or equal to the preset green signal ratio standard value, the green signal ratio of the current intersection is in accordance with the standard, the control module does not adjust the green signal ratio in the traffic signal conversion period of other intersections, and if the green signal ratio in the traffic signal conversion period of the current intersection is greater than the preset green signal ratio standard value, the control module increases the green signal ratio in the traffic signal conversion period of the next intersection, so that vehicles passing through the current intersection smoothly pass through the next intersection, the blockage of the next intersection is avoided, and meanwhile, the green signal ratio in the traffic signal conversion period of the previous intersection is shortened, and the situation that the current intersection waits for too many vehicles and is avoided, thereby causing the blockage.
Particularly, the number of vehicles passing through the current intersection in a traffic signal lamp conversion period in unit time is set as the utilization rate of green lamps at the current intersection, the control module compares the utilization rate of the green lamps at the current intersection with the preset utilization rate of the green lamps of the control module to judge whether the model constructed by the current model construction module and the pretreatment of the pretreatment module meet the standard or not for evaluation, wherein if the utilization rate of the green lamps at the current intersection is smaller than or equal to the first preset utilization rate of the green lamps, the conversion cycle of the traffic signal lamps at the current intersection is not good, the control module judges that the preset traffic signal lamp conversion cycle of the model construction unit is reduced, if the utilization rate of the green lamps at the current intersection is between the first preset utilization rate of the green lamps and the second preset utilization rate of the green lamps, the judgment on the blockage situation of the current intersection is not accurate, and the control module improves the accuracy of judging the blockage situation at the current intersection by enlarging the preset area of the preset module.
Drawings
FIG. 1 is a schematic structural diagram of an urban traffic management adaptive system based on deep learning driving according to an embodiment of the present invention;
FIG. 2 is a side view of the camera placement position of the deep learning driven urban traffic management adaptive system according to the embodiment of the present invention;
FIG. 3 is a top view of the camera placement position of the deep learning driven urban traffic management adaptive system according to the embodiment of the present invention;
fig. 4 is a schematic diagram of a deep learning network structure of the deep learning-driven urban traffic management adaptive system according to the embodiment of the present invention.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described in conjunction with the following examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.
It should be noted that in the description of the present invention, the terms of direction or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, which are only for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
Referring to fig. 1, it is a schematic structural diagram of an adaptive system for urban traffic management based on deep learning driving according to an embodiment of the present invention, including,
the model building module comprises a storage unit and a model building unit, wherein the storage unit is used for storing vehicle flow of each road junction in each time period, the model building unit is used for building a traffic signal lamp conversion period model according to the vehicle flow of each road junction stored in the storage unit, and the model building unit regulates the traffic signal lamp conversion period according to the green light utilization rate;
the acquisition module comprises a plurality of acquisition units, is arranged at each intersection and is used for acquiring images at each intersection;
the preprocessing module is connected with the acquisition module and used for preprocessing the images of the intersections acquired by the acquisition module and predicting the blockage levels of the intersections, wherein the preprocessing module acquires the information of vehicles to be passed in the preset image area of the current intersection to acquire the urgency of the current intersection and compares the acquired urgency of the current intersection with the preset urgency to predict the blockage levels of the current intersection;
and the control module is connected with the model building model, the acquisition module and the preprocessing module and is used for adjusting the green signal ratio in the current intersection traffic signal lamp conversion period and the current intersection traffic signal lamp conversion period according to the current intersection jam level, wherein if the current intersection jam level is A1 level, the control module reduces the current intersection traffic signal lamp conversion period, if the current intersection jam level is A3 level, the control module improves the green signal ratio in the current intersection traffic signal lamp conversion period, and meanwhile, the control module compares the acquired current intersection green signal ratio with a preset green signal ratio standard value and adjusts the green signal ratio in the next intersection traffic signal lamp conversion period and the previous intersection so as to enable the traffic condition of the current intersection to accord with a preset standard.
Fig. 2 is a side view of a camera placement position of an urban traffic management adaptive system based on deep learning driving according to an embodiment of the present invention, and fig. 3 is a top view of a camera placement position of an urban traffic management adaptive system based on deep learning driving according to an embodiment of the present invention.
Please refer to fig. 4, which is a schematic diagram of a deep learning network structure of an urban traffic management adaptive system based on deep learning driving according to an embodiment of the present invention, wherein a transform network structure is adopted to perform a target detection task, so as to analyze the urgency of a current intersection, and through pre-trained network parameters, the current intersection can be converged faster on a new data set, and over-fitting is prevented.
Specifically, the method for constructing the model by the model construction unit in the embodiment of the invention is not limited, and the embodiment of the invention provides a method for constructing a traffic signal lamp conversion period model, which is provided with modules for running object detection by various methods, and OpenCV calculates a vehicle through collected images, wherein OpenCV is an image processing library written by C + + and Python, calculated vehicle information is provided to a designed transform network through OpenCV, the transform is a neural network model based on a multi-head attention mechanism and used for real-time target detection, target detection is executed by the transform model, the number of two-wheel vehicles and four-wheel vehicles in a graph is calculated, and total passing time is calculated according to the number of the two-wheel vehicles and the four-wheel vehicles, so that the traffic signal lamp conversion period is regulated.
Wherein, the model construction unit presets a flow L, the model construction unit selects each time period cycle to construct a current intersection traffic light switching cycle module according to the comparison between the acquired vehicle flow lk at each time period of the current intersection and the preset flow, wherein,
when lk is less than or equal to L1, the model construction unit selects a first preset period T1 as a k-th time period traffic signal lamp conversion period;
when L1 is larger than lk and is smaller than or equal to L2, the model construction unit selects a second preset period T2 as a kth time period traffic light conversion period;
when L2 is larger than lk and smaller than or equal to L3, the model construction unit selects a third preset period T3 as a k-th time period traffic signal lamp conversion period;
when lk is larger than L3, the model construction unit selects a fourth preset period T4 as a k-th time period traffic signal lamp conversion period;
the flow rate L is preset by the model construction unit, a first preset flow rate L1, a second preset flow rate L2, a third preset flow rate L3 and a fourth preset flow rate L4 are set, the period T is preset by the model construction unit, and a first preset period T1, a second preset period T2, a third preset period T3 and a fourth preset period T4 are set, wherein k =1, 2.
Specifically, the model construction unit divides the preset period into four definite standards, the model construction unit compares the current intersection vehicle flow in each time period stored in the storage unit with the preset flow, and selects the optimal flow as the current intersection traffic light conversion period in each time period, wherein if the vehicle flow in the current time period of the current intersection acquired by the model construction unit is less than or equal to the first preset flow, the model construction module selects a smaller period as the current intersection traffic light conversion period in the current time period, and so on, the model construction module compares the acquired flow with the preset flow, and selects a corresponding period as the current time period traffic light conversion period, and the model construction unit constructs the traffic light conversion period model according to the acquired traffic light conversion period in each time period of the current intersection.
Specifically, in the embodiment of the present invention, the model construction unit is configured to set the vehicle flow of each time segment at the current intersection as the vehicle flow of each time segment at the current intersection according to the current intersection history data stored in the storage unit, the vehicle flow of each time segment is set as the average value of the vehicle flow of each time segment at the intersection according to each time segment stored in the storage unit, and the model construction module is configured to construct the full-time-segment traffic light conversion cycle model according to the acquired period of each time segment, where when the model construction module acquires the first time-segment traffic light conversion cycle T1, the second time-segment traffic light conversion cycle T2, the third time-segment traffic light conversion cycle T2, the fourth time-segment traffic light conversion cycle T3, the fifth time-segment traffic light conversion cycle T4, the sixth time-segment traffic light conversion cycle T2, the seventh time-segment traffic light conversion cycle T4, and the eighth time-segment traffic light conversion cycle T1, the model construction unit acquires the full-segment traffic light conversion cycle model as (T1, T2, T3, T4, T2, T4, T1), wherein the division of each time segment can be performed according to the traffic light division, and the traffic light division can be performed according to the congestion situation in all days.
Wherein, the preprocessing module acquires the current crossing urgency degree p according to the distance s between the last vehicle in the preset area and the first vehicle at the crossing and the vehicle running average speed v0 in the image acquired by the acquisition module, sets p = hj x (s/v 0)/Tig, compares the acquired current crossing urgency degree with the preset urgency degree, and predicts the current crossing jam grade, wherein,
when P is less than or equal to P1, the preprocessing module predicts that the current intersection blockage level is a first preset blockage level A1;
when P1 is more than P and less than P2, the preprocessing module predicts that the current intersection blockage grade is a second preset blockage grade A2;
when P is larger than or equal to P2, the preprocessing module predicts that the current intersection blockage level is a third preset blockage level A3;
the pre-processing module is used for presetting a congestion grade A, setting a first preset congestion grade A1, a second preset congestion grade A2 and a third preset congestion grade A3, hj is an urgency degree compensation parameter, and i =1,2,3,4 and Tig is the green time in the conversion period of the Ti traffic signal lamp.
Specifically, the vehicle density M is preset by the preprocessing module, the vehicle density M in the current preset area is obtained by the preprocessing module, and compared with the preset vehicle density, the urgency degree compensation parameter is selected, wherein,
when M is less than or equal to M1, the preprocessing module selects a first preset urgency degree compensation parameter h1 as a current intersection urgency degree compensation parameter;
when M1 is larger than M and smaller than M2, the preprocessing module selects a second preset urgency degree compensation parameter h2 as a current intersection urgency degree compensation parameter;
when M is larger than or equal to M2, the preprocessing module selects a third preset urgency degree compensation parameter h3 as a current intersection urgency degree compensation parameter;
the vehicle density M is preset by the preprocessing module, a first preset vehicle density M1, a second preset vehicle density M2, a preset urgency degree compensation parameter h, a first preset urgency degree compensation parameter h1, a second preset urgency degree compensation parameter h2 and a third preset urgency degree compensation parameter h3 are set.
Specifically, the method includes the steps that a preprocessing module is arranged to obtain a current intersection urgency degree according to the total length of vehicles in a preset area and the average speed of vehicle running, and a congestion level of the current intersection is obtained according to the current intersection urgency degree and the preset urgency degree, wherein the preprocessing module compares the vehicle density in the preset area with the preset vehicle density, selects an optimal urgency degree compensation parameter to compensate the urgency degree, and avoids the situation that due to the fact that the vehicle density is too low, gaps among the vehicles are short, the time of the vehicles running at the average speed of the vehicles is short, and the congestion degree is not accurately obtained.
Specifically, the embodiment of the present invention does not limit the vehicle density obtaining manner in the current preset area, and the vehicle density obtaining manner may be calculated according to a vehicle clearance, or may be calculated according to a distance between a last vehicle in the preset area and a first vehicle at an intersection and the number of vehicles, where an embodiment of the present invention provides a preferred implementation scheme, and the vehicle density is obtained according to the lengths and the number of all vehicles in the preset area, that is, the vehicle density m = s/num, where num is the total number of vehicles in the preset area, and s is the total length of vehicles in the preset area.
Specifically, the embodiment of the present invention does not limit the manner of obtaining the vehicle running speed, and the embodiment of the present invention provides a preferred implementation scheme, where the average value of the running speeds of the vehicles at the current intersection in the history record stored in the storage module is set as the vehicle running average speed to accurately obtain the urgency degree of the current intersection.
The control module adjusts the traffic light switching period and the green signal ratio in the period of the current intersection according to the congestion level of the current intersection predicted by the preprocessing module, wherein,
if the current intersection jam level is A1, the control module reduces the current intersection traffic signal lamp conversion period Ti to Ti1, and sets Ti1= Ti x (1- (P1-P)/P1);
if the current intersection jam level is A2, the control module does not adjust the current intersection traffic signal lamp conversion period and the green signal ratio in the period;
if the current intersection congestion level is A3, the control module increases the split green ratio Tig to Tig1 in the current intersection traffic light conversion period, and sets Tig1= Tig x (1 + (P-P2)/P2).
Specifically, the embodiment of the invention provides a preferable deep learning model for calculating the time for two-wheel vehicles (TW) and four-wheel vehicles (FW) to pass through a road at the same time, estimating a time period according to the number of vehicles and the time for each to pass through a lane, finally allocating a green light time t,
t＝[TW/a]×n+[FW/b]×m
where TW represents the number of two-wheeled vehicles, FW represents the number of four-wheeled vehicles, n represents the time for passing only two-wheeled vehicles alone, m represents the time for passing only four-wheeled vehicles alone, a represents the maximum number of two-wheeled vehicles all over the road surface, and b represents the maximum number of four-wheeled vehicles all over the road surface.
However, when there are a large number of vehicles at the intersection, the vehicles need to pass for a longer time, which results in a long waiting time for the vehicles in the opposite lane. Therefore, to overcome the overall drawback, we set a maximum threshold time, so the actual start time of green light is calculated as follows:
t ture ＝min(t,max_green_time)
to calculate max _ green _ time, we assume that all passing cars are four-wheel cars, so max _ green _ time is calculated as 42s, determining the maximum green light time threshold.
Specifically, the acquired green-signal ratio in the conversion period of the traffic signal lamp at the current intersection is adjusted according to the grade of the congestion of the current intersection predicted by the preprocessing module, wherein when the grade of the congestion of the current intersection predicted by the preprocessing module is A1 grade, the fact that the congestion degree of the current intersection is low is indicated, in order to improve the efficiency of traffic conversion, the control module reduces the conversion period of the traffic signal lamp at the current intersection, when the grade of the congestion of the current intersection predicted by the preprocessing module is A2, the fact that the congestion condition of the current intersection is in a standard range is indicated, the control module does not adjust the green-signal ratio in the conversion period and the current traffic signal lamp conversion period of the current intersection, when the grade of the congestion of the current intersection predicted by the preprocessing module is A3, the fact that the current intersection is about to be seriously congested is indicated, and the control module improves the green-signal ratio in the conversion period of the traffic signal lamp at the current intersection so as to solve the possible congestion problem at the current intersection.
Specifically, in the embodiment of the invention, the green ratio is that the time of turning on the green light accounts for the time of the whole traffic signal light conversion period, and the traffic signal light conversion period is that the time of turning on the green light, the time of turning on the yellow light and the time of turning on the red light are one period.
Wherein, the control module obtains the green signal ratio Tig1 in the traffic light switching period of the current intersection and compares the green signal ratio Tig1 with a preset green signal ratio standard value T0 of the control module, and adjusts the green signal ratio in the traffic light switching period of the previous intersection and the next intersection, wherein,
when the Tig1 is less than or equal to T0, the control module does not adjust the split ratio of the traffic signal lamps at other intersections in the conversion period;
when the Tig1 is larger than T0, the control module improves the split ratio in the traffic signal lamp conversion period of the next intersection and shortens the split ratio in the traffic signal lamp conversion period of the previous intersection.
Specifically, the control module is provided with a green signal ratio standard value, and compares the obtained green signal ratio in the traffic signal conversion period of the current intersection with a preset green signal ratio standard value to adjust the green signal ratio of other intersections, wherein if the green signal ratio in the traffic signal conversion period of the current intersection is less than or equal to the preset green signal ratio standard value, the green signal ratio of the current intersection is in accordance with the standard, the control module does not adjust the green signal ratio in the traffic signal conversion period of other intersections, and if the green signal ratio in the traffic signal conversion period of the current intersection is greater than the preset green ratio standard value, the control module increases the green signal ratio in the traffic signal conversion period of the next intersection, so that vehicles passing through the current intersection can smoothly pass through the next intersection, the next intersection is prevented from being blocked, and meanwhile, the green signal ratio in the traffic signal conversion period of the previous intersection is shortened, and the current intersection is prevented from waiting for too many vehicles to cause blockage.
The control module obtains that the split ratio in the traffic signal light conversion period of the current intersection is larger than a preset split ratio standard value, the control module increases the split ratio Tx to Tx1 in the traffic signal light conversion period of the next intersection, sets Tx1= Txx (1 + (Tig 1-T0)/T0), shortens the split ratio Ts to Ts1 in the traffic signal light conversion period of the previous intersection, and sets Ts1= Ts x (1- (Tig 1-T0)/T0).
Specifically, the control module obtains a current intersection green light utilization rate F, sets F = d/Ti, wherein d is the number of vehicles passing through the current intersection in the current traffic light conversion period, compares the obtained current intersection green light utilization rate with a preset green light utilization rate F0, and adjusts a preset area of the preprocessing module and the preset traffic light conversion period of the model building unit, wherein,
when F is not more than F1, the control module reduces a preset traffic signal lamp conversion period Ti to Ti2 of the model construction unit, and sets Ti2= Ti x (1- (F1-F)/F1);
when the F1 is larger than the F and smaller than the F1, the control module does not adjust the preset area;
when F is larger than or equal to F2, the control module enlarges a preset area of the preprocessing module;
the control module presets a green light utilization rate F, sets a first preset green light utilization rate F1, and presets a second green light utilization rate F2.
Specifically, when the control module obtains that the green light utilization rate of the current intersection is greater than or equal to a second preset green light utilization rate, the control module enlarges a preset area w to w1 of the preprocessing module, and sets w1= w × (1 + (F-F2)/F2).
Specifically, the preset region set by the preprocessing module in the embodiment of the present invention is not limited, as long as the preset region can intercept and analyze the image acquired by the acquisition module, and the embodiment of the present invention provides an optimal implementation scheme, that is, the preset region is an image length ratio, more specifically, the preset region in the embodiment of the present invention is a preprocessing module processing region with half of the length of the image, and when the control module adjusts the preset region, the adjusted preset region occupies 2/3 of the image, that is, 2/3 of the image is a preprocessing module processing region.
Specifically, the number of vehicles passing through the current intersection in a traffic light conversion period in unit time is set as the current intersection green light utilization rate, the control module compares the current intersection green light utilization rate with the preset green light utilization rate of the control module to judge whether the model constructed by the current model construction module and pretreatment of the pretreatment module meet standards or not to evaluate, wherein if the current intersection green light utilization rate is smaller than or equal to the first preset green light utilization rate, the current intersection traffic light conversion period is not good, the control module judges to reduce the preset traffic light conversion period of the model construction unit, if the current intersection green light utilization rate is between the first preset green light utilization rate and the second preset green light utilization rate, the current intersection passing vehicles meet the preset standards, the control module does not adjust the preset area of the pretreatment module and the preset traffic light conversion period of the model construction unit, if the current intersection green light utilization rate is larger than or equal to the second preset green light utilization rate, the current intersection jam judgment is not accurate, and the control module improves the accuracy of judging the current intersection jam situation by enlarging the preset area of the pretreatment module.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is apparent to those skilled in the art that the scope of the present invention is not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (10)
1. An adaptive system for urban traffic management based on deep learning drive, comprising:
the model building module comprises a storage unit and a model building unit, wherein the storage unit is used for storing vehicle flow of each road junction in each time period, the model building unit is used for building a traffic signal lamp conversion period model according to the vehicle flow of each road junction stored in the storage unit, and the model building unit regulates the traffic signal lamp conversion period according to the green light utilization rate;
the acquisition module comprises a plurality of acquisition units, is arranged at each intersection and is used for acquiring images at each intersection;
the preprocessing module is connected with the acquisition module and used for preprocessing the images of the intersections acquired by the acquisition module and predicting the blockage levels of the intersections, wherein the preprocessing module acquires the information of vehicles to be passed in the preset image area of the current intersection to acquire the urgency of the current intersection and compares the acquired urgency of the current intersection with the preset urgency to predict the blockage levels of the current intersection;
and the control module is connected with the model building model, the acquisition module and the preprocessing module and is used for adjusting the green signal ratio in the current intersection traffic signal lamp conversion period and the current intersection traffic signal lamp conversion period according to the current intersection jam level, wherein if the current intersection jam level is A1 level, the control module reduces the current intersection traffic signal lamp conversion period, if the current intersection jam level is A3 level, the control module improves the green signal ratio in the current intersection traffic signal lamp conversion period, and meanwhile, the control module compares the acquired current intersection green signal ratio with a preset green signal ratio standard value and adjusts the green signal ratio in the next intersection traffic signal lamp conversion period and the previous intersection so as to enable the traffic condition of the current intersection to accord with a preset standard.
2. The deep learning driven urban traffic management adaptive system according to claim 1, wherein the model construction unit presets a flow rate L, the model construction unit selects each time period cycle to construct a current intersection traffic light switching cycle module according to the comparison between the acquired vehicle flow rate lk and the preset flow rate at each time period of the current intersection, wherein,
when lk is less than or equal to L1, the model construction unit selects a first preset period T1 as a k-th time period traffic signal lamp conversion period;
when L1 is larger than lk and is smaller than or equal to L2, the model construction unit selects a second preset period T2 as a kth time period traffic light conversion period;
when the Lk is more than L2 and less than or equal to L3, the model construction unit selects a third preset period T3 as a kth time period traffic light conversion period;
when lk is larger than L3, the model construction unit selects a fourth preset period T4 as a kth time period traffic light conversion period;
the flow L is preset by the model building unit, a first preset flow L1, a second preset flow L2, a third preset flow L3 and a fourth preset flow L4 are set, the period T is preset by the model building unit, and a first preset period T1, a second preset period T2, a third preset period T3 and a fourth preset period T4 are set, k =1,2,. N and n are the total time period.
3. The deep learning driven urban traffic management adaptive system according to claim 2, wherein the preprocessing module obtains a current intersection urgency p according to a distance s between a last vehicle in a preset area and a first vehicle at an intersection in the image obtained by the acquisition module and a vehicle running average speed v0, and sets p = hj x (s/v 0)/Tig, the preprocessing module compares the obtained current intersection urgency with a preset urgency, and the preprocessing module predicts a current intersection congestion level, wherein,
when P is less than or equal to P1, the preprocessing module predicts that the current intersection blockage level is a first preset blockage level A1;
when P1 is larger than P and smaller than P2, the preprocessing module predicts that the current intersection blockage grade is a second preset blockage grade A2;
when P is larger than or equal to P2, the preprocessing module predicts that the current intersection blockage grade is a third preset blockage grade A3;
the pre-processing module is used for presetting a congestion grade A, setting a first preset congestion grade A1, a second preset congestion grade A2 and a third preset congestion grade A3, hj is an urgency degree compensation parameter, and i =1,2,3,4 and Tig is the green time in the conversion period of the Ti traffic signal lamp.
4. The deep learning driven urban traffic management adaptive system according to claim 3, wherein the pre-processing module presets a vehicle density M, acquires a vehicle density M in a current preset area, compares the vehicle density M with the preset vehicle density, and selects a urgency degree compensation parameter, wherein,
when M is less than or equal to M1, the preprocessing module selects a first preset urgency degree compensation parameter h1 as a current intersection urgency degree compensation parameter;
when M1 is larger than M and smaller than M2, the preprocessing module selects a second preset urgency degree compensation parameter h2 as a current intersection urgency degree compensation parameter;
when M is larger than or equal to M2, the preprocessing module selects a third preset urgency degree compensation parameter h3 as a current intersection urgency degree compensation parameter;
the vehicle density M is preset by the preprocessing module, a first preset vehicle density M1, a second preset vehicle density M2, a preset tightness compensation parameter h, a first preset tightness compensation parameter h1, a second preset tightness compensation parameter h2 and a third preset tightness compensation parameter h3 are set.
5. The adaptive system for urban traffic management based on deep learning driving of claim 3, wherein the control module adjusts the turn-over period and the green signal ratio in the turn-over period of the traffic signal lamp at the current intersection according to the congestion level predicted by the preprocessing module at the current intersection,
if the current intersection congestion level is A1, the control module reduces the current intersection traffic light switching period Ti to Ti1, and sets Ti1= Ti x (1- (P1-P)/P1);
if the current intersection jam level is A2, the control module does not adjust the current intersection traffic signal lamp conversion period and the green signal ratio in the period;
if the current intersection jam level is A3, the control module increases the split ratio Tig in the current intersection traffic light conversion period to Tig1, and Tig1= Tig x (1 + (P-P2)/P2) is set.
6. The deep learning driven urban traffic management adaptive system according to claim 5, wherein the control module obtains the split Tig1 of the current intersection in the traffic light switching period and compares the split Tig1 with the preset split standard value T0 of the control module to adjust the split of the previous intersection and the next intersection in the traffic light switching period, wherein,
when the Tig1 is less than or equal to T0, the control module does not adjust the split ratio of the traffic signal lamps at other intersections in the conversion period;
when Tig1 is larger than T0, the control module improves the split ratio in the traffic signal lamp conversion period of the next intersection and shortens the split ratio in the traffic signal lamp conversion period of the previous intersection.
7. The urban traffic management adaptive system based on deep learning driving according to claim 6, wherein the control module obtains that the split ratio in the traffic light switching period of the current intersection is greater than a preset split ratio standard value, the control module increases the split ratio Tx to Tx1 in the traffic light switching period of the next intersection, sets Tx1= Tx (1 + (Tig 1-T0)/T0), shortens the split ratio Ts to Ts1 in the traffic light switching period of the previous intersection, and sets Ts1= Ts x (1- (Tig 1-T0)/T0).
8. The deep learning driven-based urban traffic management adaptive system according to claim 7, wherein the control module compares the obtained current intersection green light utilization rate F with a preset green light utilization rate F0, and adjusts a preset area of the preprocessing module and a preset traffic signal light switching period of the model building unit, wherein,
when F is not more than F1, the control module reduces a preset traffic signal lamp conversion period Ti to Ti2 of the model construction unit, and sets Ti2= Ti x (1- (F1-F)/F1);
when the F1 is larger than the F and smaller than the F1, the control module does not adjust the preset area;
when F is larger than or equal to F2, the control module expands the preset area of the preprocessing module;
the control module presets a green light utilization rate F, sets a first preset green light utilization rate F1, and presets a second green light utilization rate F2.
9. The deep learning driven-based urban traffic management adaptive system according to claim 8, wherein the control module obtains a green light utilization rate f at the current intersection, and sets f = d/Ti, wherein d is the number of vehicles passing through the current intersection in the current traffic light conversion period.
10. The deep learning driven urban traffic management adaptive system according to claim 9, wherein when the control module obtains that the green light utilization rate at the current intersection is greater than or equal to a second preset green light utilization rate, the control module enlarges the preset area w to w1 of the preprocessing module, and sets w1= w x (1 + (F-F2)/F2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211178762.1A CN115662152B (en) | 2022-09-27 | 2022-09-27 | Urban traffic management self-adaptive system based on deep learning driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211178762.1A CN115662152B (en) | 2022-09-27 | 2022-09-27 | Urban traffic management self-adaptive system based on deep learning driving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115662152A true CN115662152A (en) | 2023-01-31 |
| CN115662152B CN115662152B (en) | 2023-07-25 |
Family
ID=84985977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211178762.1A Active CN115662152B (en) | 2022-09-27 | 2022-09-27 | Urban traffic management self-adaptive system based on deep learning driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115662152B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103093633A (en) * | 2011-10-28 | 2013-05-08 | 国际商业机器公司 | Adjustment system and method of traffic signal lamps |
| JP2017016554A (en) * | 2015-07-06 | 2017-01-19 | 株式会社日立製作所 | Traffic light control system and traffic light control method |
| CN107358803A (en) * | 2017-08-22 | 2017-11-17 | 哈尔滨理工大学 | A kind of traffic signal control system and its control method |
| CN108417057A (en) * | 2018-05-15 | 2018-08-17 | 哈尔滨工业大学 | A kind of intelligent signal lamp timing system |
| CN109615884A (en) * | 2018-12-20 | 2019-04-12 | 南京丹腾智能科技有限公司 | Wireless traffic signal light management system |
| CN111613070A (en) * | 2019-02-25 | 2020-09-01 | 北京嘀嘀无限科技发展有限公司 | Traffic signal lamp control method, traffic signal lamp control device, electronic equipment and computer storage medium |
| CN112863198A (en) * | 2020-12-29 | 2021-05-28 | 连云港杰瑞电子有限公司 | Signal lamp green signal ratio intelligent adjustment method for expressway entrance ramp |
| CN112991750A (en) * | 2021-05-14 | 2021-06-18 | 苏州博宇鑫交通科技有限公司 | Local traffic optimization method based on reinforcement learning and generation type countermeasure network |
| CN113096420A (en) * | 2021-03-31 | 2021-07-09 | 联想(北京)有限公司 | Information processing method, device and system |
| CN113192345A (en) * | 2021-04-12 | 2021-07-30 | 武汉理工大学 | Traffic light intelligent regulation and control system for crossroad |
| CN113380054A (en) * | 2021-06-09 | 2021-09-10 | 湖南大学 | Traffic signal lamp control method and system based on reinforcement learning |
| US20210358298A1 (en) * | 2020-05-14 | 2021-11-18 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for adjusting channelization of traffic intersection |
| CN114141028A (en) * | 2021-11-19 | 2022-03-04 | 哈尔滨工业大学（深圳） | Intelligent traffic light traffic flow regulation and control system |
-
2022
- 2022-09-27 CN CN202211178762.1A patent/CN115662152B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103093633A (en) * | 2011-10-28 | 2013-05-08 | 国际商业机器公司 | Adjustment system and method of traffic signal lamps |
| JP2017016554A (en) * | 2015-07-06 | 2017-01-19 | 株式会社日立製作所 | Traffic light control system and traffic light control method |
| CN107358803A (en) * | 2017-08-22 | 2017-11-17 | 哈尔滨理工大学 | A kind of traffic signal control system and its control method |
| CN108417057A (en) * | 2018-05-15 | 2018-08-17 | 哈尔滨工业大学 | A kind of intelligent signal lamp timing system |
| CN109615884A (en) * | 2018-12-20 | 2019-04-12 | 南京丹腾智能科技有限公司 | Wireless traffic signal light management system |
| CN111613070A (en) * | 2019-02-25 | 2020-09-01 | 北京嘀嘀无限科技发展有限公司 | Traffic signal lamp control method, traffic signal lamp control device, electronic equipment and computer storage medium |
| US20210358298A1 (en) * | 2020-05-14 | 2021-11-18 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for adjusting channelization of traffic intersection |
| CN112863198A (en) * | 2020-12-29 | 2021-05-28 | 连云港杰瑞电子有限公司 | Signal lamp green signal ratio intelligent adjustment method for expressway entrance ramp |
| CN113096420A (en) * | 2021-03-31 | 2021-07-09 | 联想(北京)有限公司 | Information processing method, device and system |
| CN113192345A (en) * | 2021-04-12 | 2021-07-30 | 武汉理工大学 | Traffic light intelligent regulation and control system for crossroad |
| CN112991750A (en) * | 2021-05-14 | 2021-06-18 | 苏州博宇鑫交通科技有限公司 | Local traffic optimization method based on reinforcement learning and generation type countermeasure network |
| CN113380054A (en) * | 2021-06-09 | 2021-09-10 | 湖南大学 | Traffic signal lamp control method and system based on reinforcement learning |
| CN114141028A (en) * | 2021-11-19 | 2022-03-04 | 哈尔滨工业大学（深圳） | Intelligent traffic light traffic flow regulation and control system |
Non-Patent Citations (1)
| Title |
|---|
| 朱文兴,贾磊,吴晓晴: "城市主干路交通流多目标优化控制" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115662152B (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114664078B (en) | Road confluence area cooperation convergence control method based on automatic driving vehicle queue | |
| CN108665707B (en) | Expressway entrance diversion and flow cooperative control method for preventing local traffic jam | |
| CN109598950B (en) | Ramp cooperative convergence control method and system for intelligent networked vehicles | |
| CN107067764B (en) | Self-adaptive control method for variable guide lane of urban intersection | |
| CN114418468A (en) | Smart city traffic scheduling strategy control method and Internet of things system | |
| CN114141021B (en) | Smart city traffic signal lamp control system and control method thereof | |
| CN107204120B (en) | Method and device for designing traffic signal phase by using travel time | |
| CN110992712A (en) | Traffic signal lamp light control system based on cloud computing | |
| US20240046787A1 (en) | Method And System For Traffic Clearance At Signalized Intersections Based On Lidar And Trajectory Prediction | |
| CN111341107A (en) | Shared traffic control method based on cloud platform data | |
| CN115206115A (en) | Road network congestion area control method based on multi-source data edge calculation in intelligent networking environment | |
| CN114360264A (en) | Intelligent city traffic management method based on traffic flow regulation | |
| CN111754790B (en) | Ramp entrance traffic control system and method based on radar | |
| CN115662152A (en) | Urban traffic management self-adaptive system based on deep learning drive | |
| CN114973709A (en) | Intelligent control system and control method for urban traffic lights | |
| CN111968397B (en) | Automatic driving behavior determination method | |
| CN115762195A (en) | Dynamic green wave signal control method, system, device and storage medium | |
| CN115527382A (en) | Traffic control method based on intelligent multi-objective optimization algorithm | |
| CN113115203B (en) | Consumption reduction method, device and system for road side unit equipment | |
| CN111489570A (en) | Traffic guidance system based on time-interval traffic flow | |
| CN114283580A (en) | Traffic flow guiding method based on big data | |
| CN109243186B (en) | Method for dynamically adjusting phase difference based on queuing level | |
| CN112509341A (en) | Signal control device and method for analyzing traffic conflicts based on high-level video | |
| CN113140105A (en) | Method and system for managing operation and scheduling of high-peak bus | |
| CN117116063B (en) | Method for detecting and adjusting traffic jam |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |