Smart city management system and method

Abstract

The invention belongs to the technical field of urban traffic management, and particularly relates to a smart city management system and method, and the system accurately collects data in real time through a current road data acquisition module and a scattered field data acquisition module, and transmits the data to a data prediction module. The data prediction module carries out analysis and prediction based on the data, and provides a key decision basis for the traffic light control module. And the traffic light control module accurately controls the switching time of the intersection signal lights according to the prediction result, so that effective dredging of a large amount of people and traffic flows in a short time is realized, and the traffic jam condition of the target intersection in the short time occurring after the commercial activity or other events in the commercial center area is relieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban traffic management, and particularly relates to a smart city management system and method. Background Art

[0002] With the acceleration of the urbanization process, the urban population density has been continuously increasing, and the vehicle ownership has been rising continuously, resulting in unprecedented pressure on the urban traffic system. Especially in high-density areas such as the Central Business District (CBD), there is a certain temporality, and the phenomenon of a sharp increase in the traffic flow and pedestrian flow within a short period is particularly prominent. Such short-term peak flows will not only cause a serious impact on the road traffic capacity. In particular, the core intersections within the business district usually refer to the intersections with the largest traffic volume, the highest pedestrian density, or the most critical geographical location, but may also trigger a chain reaction, further exacerbating the traffic congestion on the roads near the commercial center. For example, during holidays or large commercial activities, the traffic volume of pedestrians and vehicles at the entrances and exits of the business district may suddenly increase several times at the end of the commercial activity or during holidays, causing traffic congestion. However, the existing traffic management models usually make predictions and scheduling based on fixed rules and historical data, and it is difficult to flexibly cope with such sudden traffic fluctuations. Specifically, between different traffic facilities, such as in a multi-intersection intersection, there is a lack of effective scheduling between multiple intersections, and there is a lack of effective linkage in the coordination and cooperation between the traffic sections around a business district, resulting in information islands and limiting the comprehensive decision-making ability.

[0003] In addition, the CBD areas of modern cities often adopt a highly integrated building structure, integrating multiple functions such as office, commerce, and residence. Although this design improves the land use efficiency, it also makes the flow of people and vehicles in the area more concentrated, further deteriorating the traffic conditions. Especially during the morning and evening rush hours, a large number of commuters pour into or leave the CBD area, forming a significant tidal effect, which poses a huge challenge to traffic management. Summary of the Invention

[0004] To solve the above problems existing in the prior art, the present invention provides a smart city management system and method, which solves the problem of congestion at the target exit caused by a large number of people and vehicles within a short period after the end of large-scale activities or holidays in the business district.

[0005] The object of the present invention can be achieved by the following technical solutions: A smart city management system, including;

[0006] A current road data acquisition module, configured to acquire real-time pedestrian density data on the current sidewalk and real-time traffic flow data on the current road;

[0007] A dispersal data acquisition module, configured to acquire dispersal vehicle data, dispersal pedestrian data, and road data from the dispersal location to the target intersection;

[0008] The data prediction module is used to predict the time when the pedestrians leaving the venue reach the target intersection based on the data of the pedestrians leaving the venue and the road data from the leaving position to the target intersection, and then combine the real-time pedestrian density data to predict the time when the pedestrian peak appears at the target intersection and the pedestrian peak value.

[0009] The traffic light control module is used to control the current intersection to be a green light for pedestrians when the time when the pedestrian peak appears arrives, and is also used to calculate the green light time for pedestrians required for the number of pedestrians at the pedestrian peak to pass through the target intersection according to the pedestrian peak and the preset traffic volume of the current intersection, and control the lighting time of the green light for pedestrians according to the calculated green light time for pedestrians.

[0010] Until after the predicted value is less than the pedestrian peak, the traffic light control module controls the current intersection to be a green light for vehicles.

[0011] The traffic light control module calculates the vehicle green light time required for the total traffic flow data to pass through the target intersection according to the total traffic flow data and the preset traffic flow passing data of the target intersection, and controls the lighting time of the vehicle green light with the vehicle green light time. The total traffic flow data includes real-time traffic flow data and departing traffic flow data.

[0012] Preferably, the data prediction module includes a traffic flow prediction unit, and the traffic flow prediction unit is used to predict the time when the vehicle peak appears at the target intersection according to the departing vehicle data and the road data from the leaving position to the target intersection, and combine the real-time traffic flow data, and transmit the time when the vehicle peak appears to the traffic light control module.

[0013] If the time when the vehicle peak appears is earlier than the end time of the green light time for pedestrians, the traffic light control module controls the target intersection to be a green light for vehicles when the time when the vehicle peak appears arrives.

[0014] Preferably, the data prediction module further includes a topology data acquisition unit, a cluster segmentation unit and a calculation unit:

[0015] The topology data acquisition unit is used to obtain the leaving position coordinates, the target intersection coordinates and multiple paths between the leaving position and the target intersection from the GIS road network data.

[0016] The cluster segmentation unit is used to divide the pedestrians on the multiple paths into multiple flowing local density clusters respectively through the DBSCAN clustering algorithm.

[0017] The calculation unit is used to monitor the moving speeds of the multiple clusters respectively according to the clusters segmented by the cluster segmentation unit by using the target tracking algorithm, and calculate the time for the multiple clusters to reach the target intersection respectively in combination with the distances from the real-time positions of the multiple clusters to the target intersection.

[0018] Preferably, the calculation unit calculates the time for multiple cluster groups to reach the target intersection based on the moving speeds of the multiple cluster groups and the distances between the locations of specific cluster groups and the target intersection, and generates a pedestrian-time curve for reaching the target intersection in combination with the number of pedestrians in each cluster group;

[0019] The calculation unit fuses the real-time pedestrian density data with the pedestrian-time curve to obtain a pedestrian-time curve for reaching the current intersection, and calculates the peak value of this curve, calibrates this peak value as the pedestrian peak value, and the time to reach this peak value as the pedestrian peak appearance time.

[0020] Preferably, the data prediction module can use the DTW algorithm to find the best time alignment path between the time series of two sets of data, including the following steps:

[0021] a: Organize the path topology data obtained from the GIS road network and the pedestrian data collected by cameras or sensors into time series respectively, and calculate the distance between each data point in the two time series using the Euclidean distance;

[0022] b: Construct an n×m distance matrix D;

[0023] c: Initialize the cumulative cost matrix C, and the cumulative cost matrix C is (n + 1)×(m + 1) - dimensional:

[0024] d: Traverse each element Ci,j of the matrix, where i corresponds to row A and j corresponds to column B, and fill the cumulative cost matrix C row by row and column by column:

[0025] e: Backtrack to find the best alignment matrix, and based on the best alignment matrix, find the best alignment path, and align the time series of the two sets of data according to the best alignment path.

[0026] Preferably, in step S4, the following formula is used to calculate the minimum cumulative cost:

[0027] C(i, j) = D(i - 1, j - 1) + min{C(i - 1, j), C(i, j - 1), C(i - 1, j - 1)};

[0028] Among them, C(i, j - 1) indicates that the time point in column B is ahead, C(i - 1, j) indicates that the time point in row A is ahead, C(i - 1, j - 1) indicates that the time points in row A and column B are synchronized, and min{C(i - 1, j), C(i, j - 1), C(i - 1, j - 1)} represents the minimum cumulative cost of the previous three paths, i = 1, 2,..., n, j = 1, 2,..., m.

[0029] Preferably, in step S5, two said optimal alignment paths are obtained by starting from the Cn,m element of the cumulative cost matrix C and backtracking to the starting point C0,0 to end.

[0030] Preferably, the road data acquisition module includes an infrared induction pedestrian counter, a geomagnetic sensor, and a high-definition camera.

[0031] A method for a smart city management system includes the following steps:

[0032] S1: Predict the time for the dispersing pedestrians to reach the target intersection by combining the dispersing pedestrian data and the road data from the dispersing location to the target intersection, and predict the time when the pedestrian peak appears and the peak data at the target intersection by combining the real-time pedestrian density data. When the time of the pedestrian flow peak data appears, control the current intersection to be a green light for pedestrians;

[0033] S2: Calculate the pedestrian green light time required for the pedestrians with the peak data quantity to pass through the target intersection according to the peak data and the preset traffic volume of the current intersection, and control the continuous time of the pedestrian green light with the said pedestrian green light time;

[0034] S3: Control the current intersection to be a green light for vehicles until the predicted value is less than the traffic peak data. The beneficial effects of the present invention are:

[0035] During the whole process of controlling the traffic lights at the target intersection according to the vehicle flow and pedestrian flow after the end of a large-scale event, each module cooperates closely. The current road data acquisition module and the dispersing data acquisition module collect data in real time and accurately, and transmit the data to the data prediction module. The data prediction module analyzes and predicts based on these data, providing key decision-making basis for the traffic light control module. The traffic light control module accurately controls the switching time of the intersection signal lights according to the prediction results, effectively dredging a large number of pedestrians and vehicles in a short time, thereby alleviating the traffic congestion at the target intersection in a short time after the end of commercial activities or other events in the commercial center area. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings.

[0037] Figure 1 It is the system structure block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following describes in detail the specific embodiments, structures, features, and effects of the present invention with reference to the accompanying drawings and preferred embodiments.

[0039] In high-density areas such as the central business district (CBD) of a city, the traffic situation changes rapidly. During special periods such as large-scale commercial activities and holidays, a large number of pedestrians and vehicles pour in or disperse within a short period of time, posing great challenges to traffic management.

[0040] Please refer to Figure 1 , this embodiment provides a management system for a smart city, including a current road data acquisition module, a crowd dispersal data acquisition module, a data prediction module, and a traffic light control module;

[0041] The current road data acquisition module is used to acquire the real-time pedestrian density data on the current sidewalk and the real-time traffic flow data on the current road. The current road data acquisition module includes an infrared induction pedestrian counter, a geomagnetic sensor, and a high-definition camera. On the sidewalk leading to the target intersection, infrared induction pedestrian counters are set at regular intervals to accurately detect the number of pedestrians passing by, and then calculate the real-time pedestrian density data. On the roadway, a combination of a geomagnetic sensor and a high-definition camera is used. The geomagnetic sensor can detect the geomagnetic changes caused by the passing of vehicles to count the number of vehicles and their approximate speeds; the high-definition camera uses image recognition technology to accurately identify vehicle types, license plate numbers, etc., and at the same time assist in counting the traffic flow to ensure accurate acquisition of the real-time traffic flow data of the vehicles that will pass through the target intersection.

[0042] Since large-scale commercial activities are often held in the central business district or during holidays, etc., after the commercial activities end or at specific times during holidays, such as zero o'clock on New Year's Day, after these specific times, a large number of people in the central business district will disperse, causing traffic congestion at the target intersection. Here, the target intersection refers to an intersection with traffic lights in the central business district.

[0043] The crowd dispersal data acquisition module is used to acquire the crowd dispersal vehicle data, the crowd dispersal pedestrian data, and the road data from the crowd dispersal location to the target intersection. Special sensors and cameras are installed at the entrances and exits of crowd dispersal locations such as event venues and shopping malls to collect the required data.

[0044] In terms of vehicles, a license plate recognition system is used to record information such as the departure time and vehicle type of vehicles. At the same time, in combination with the parking lot management system, data such as the parking duration of vehicles is obtained, and thus the data of vehicles leaving the venue is comprehensively formed. For pedestrians leaving the venue, infrared induction counting devices and face recognition cameras set at the entrance and exit are used to count information such as the number of pedestrians and the moving direction of pedestrians. When obtaining the road data from the leaving position to the target intersection, a Geographic Information System (GIS) and a real-time traffic condition monitoring system are used. The GIS system provides basic information about the road, such as road length, number of lanes, slope, etc.; the real-time traffic condition monitoring system obtains dynamic information such as road congestion and average vehicle speed in real time through sensors, cameras, and traffic flow monitoring devices installed on the road. For example, after a large-scale event is held in a shopping mall, the devices at the entrance and exit of the shopping mall quickly collect data on vehicles and pedestrians leaving the venue. At the same time, the GIS and real-time traffic condition monitoring systems provide road data from the shopping mall to the surrounding core intersections, including information such as a certain road being narrowed due to construction and the passing speed being reduced.

[0045] The currently obtained road data acquisition module and the leaving venue data acquisition module respectively transmit the obtained data to the data prediction module.

[0046] The data prediction module uses data analysis algorithms and models to integrate and analyze the leaving venue pedestrian data and the road data from the leaving position to the target intersection, predicts the time when the leaving venue pedestrians reach the target intersection with the leaving venue pedestrian data and the road data from the leaving position to the target intersection, and then combines the real-time pedestrian density data to predict the time when the peak of pedestrians reaching the target intersection appears and the peak data.

[0047] First, according to information such as the number of leaving venue pedestrians, moving direction, traffic capacity of the road, and congestion situation, a path planning algorithm and a time prediction model are used to predict the time when the leaving venue pedestrians reach the target intersection. For example, after a shopping mall event, the number of leaving venue pedestrians is 5,000. The only road leading to the target intersection is 1 kilometer long. The current road is congested, the average vehicle speed is 10 kilometers per hour, the average walking speed of pedestrians is 5 kilometers per hour, and the effective width of the main road that can accommodate pedestrian passage is limited. Through model calculation, it can be obtained that this batch of leaving venue pedestrians will take about 30 minutes to reach the target intersection.

[0048] Then, the predicted arrival time of the leaving venue pedestrians is combined with the real-time pedestrian density data provided by the currently obtained road data acquisition module, and a time series analysis model and machine learning algorithms are used to predict the time when the peak of pedestrians reaching the target intersection appears and the peak data.

[0049] The data prediction module transmits the predicted time when the peak of pedestrians at the target intersection appears and the pedestrian peak to the traffic light control module.

[0050] The traffic light control module controls the current intersection to be a green light for pedestrians when the time of pedestrian peak arrival arrives;

[0051] The traffic light control module also calculates the green light time for pedestrians required for the number of pedestrians with peak data to pass through the target intersection based on the peak data and the preset traffic volume of the current intersection, and controls the continuous duration of the green light for pedestrians with the green light time for pedestrians. Here, the preset traffic volume of the current intersection is the planned pedestrian flow data per unit time during the construction of the target intersection;

[0052] After the peak data, the traffic light control module controls the current intersection to be a green light for vehicles;

[0053] The traffic light control module calculates the green light time for vehicles required for the total traffic flow data to pass through the target intersection based on the total traffic flow data and the preset traffic flow passing data of the target intersection, and controls the continuation of the green light for vehicles with the green light time for vehicles. The total traffic flow data includes real-time traffic flow data and departing traffic flow data. Here, the preset traffic flow passing data of the target intersection is the planned traffic flow data per unit time during the construction of the target intersection.

[0054] During the entire process of controlling the traffic lights at the target intersection according to the traffic flow and pedestrian flow after the end of a large-scale event, each module cooperates closely. The current road data acquisition module and the end-of-event data acquisition module collect data in real time and accurately, and transmit the data to the data prediction module. The data prediction module analyzes and predicts based on these data, providing key decision-making basis for the traffic light control module. The traffic light control module accurately controls the switching time of the intersection signal lights according to the prediction results, effectively dredging a large number of pedestrians and vehicles in a short period of time, thereby alleviating the traffic congestion at the target intersection in a short period of time after the end of commercial activities or other events in the commercial center area.

[0055] After the end of commercial activities, a large number of pedestrians pour out from places such as shopping malls. According to prediction and calculation, the green light time for pedestrians is relatively long. If the previous mode of switching to a green light for vehicles after the end of the green light for pedestrians is still followed, during the green light period for pedestrians, the vehicles at the target intersection are in a queuing waiting state. When the vehicles arrive concentratedly later, they gradually reach the vehicle peak state, which will cause congestion in the vehicle queue at the target intersection and further affect the traffic fluency at the target intersection. Therefore, in one embodiment, the data prediction module further includes a traffic flow prediction unit. The traffic flow prediction unit is used to predict the vehicle peak arrival time at the target intersection based on the departing vehicle data, the road data from the end-of-event location to the target intersection, and in combination with the real-time traffic flow data. The traffic flow prediction unit transmits the predicted vehicle peak arrival time at the target intersection to the traffic light control module;

[0056] If the vehicle peak time appears earlier than the end time of the pedestrian green light, when the vehicle peak time arrives, the traffic light control module responds quickly. When the vehicle peak time arrives, it controls the target intersection to have a green light for vehicles. For example, the pedestrian green light time is calculated to be 60 seconds, while the predicted vehicle peak time is 45 seconds. When the 45th second arrives, the traffic light control module immediately controls the target intersection to have a green light for vehicles to clear a part of the vehicles first.

[0057] After switching to a green light for vehicles, the traffic light control module recalculates and adjusts the green light time for vehicles according to the total traffic flow data (real-time traffic flow data and departing traffic flow data) and the preset traffic flow passing data of the target intersection. For example, it is known that there are 3 lanes on the roadway of the current intersection, the passing capacity of each lane per hour is 1200 vehicles, the real-time traffic flow is 2500 vehicles per hour, and the expected departing traffic flow is 1500 vehicles per hour. After calculation, it is obtained that the green light for vehicles needs to last for 200 seconds to ensure that vehicles can pass through the intersection orderly during the vehicle peak period.

[0058] By predicting the vehicle peak time through the data prediction module and comparing it with the pedestrian green light time, the traffic light control module switches to a green light for vehicles in a timely manner when the vehicle peak arrives, avoiding the waste of vehicle passing resources during the pedestrian green light period. Giving the vehicle a green light time during the vehicle peak period ensures that vehicles can pass through the intersection quickly, reducing vehicle backlogs and congestion. The method of dynamically adjusting the signal light time according to real-time traffic data effectively solves the problem of low traffic efficiency caused by the mismatch between the pedestrian green light time and the vehicle peak time, improves the utilization rate of road resources, and alleviates the traffic congestion situation in high-density urban areas during special periods.

[0059] Before the peak pedestrian flow arrives, there are relatively few pedestrians on the road and the vehicle passing conditions are relatively good. At this time, extending the green light time for vehicles can allow more vehicles to pass through the intersection smoothly during this period, improving the vehicle passing capacity of the road. Therefore, in one embodiment, the traffic light control module extends the green light time for vehicles before the peak pedestrian flow arrives. Extending the green light time for vehicles in advance can, to a certain extent, reduce the traffic pressure caused by the subsequent restriction of vehicle passing due to the increase in pedestrian flow, avoid excessive backlogs of vehicles at the intersection and surrounding roads, and thus alleviate the traffic congestion situation in the entire area. The specific extension time is predicted in combination with specific models and real-time data.

[0060] In one embodiment, the data prediction module includes a topology data acquisition unit, a cluster segmentation unit, and a calculation unit;

[0061] The topological data acquisition unit is used to obtain the evacuation location coordinates, the coordinates of the target intersection, and multiple paths between the evacuation location and the target intersection from the GIS road network data, and construct a path network from the evacuation location to the target intersection based on the multiple paths;

[0062] Input the evacuation pedestrian data and the road data from the evacuation location to the target intersection into the DBSCAN algorithm to identify the groups of multiple paths for the evacuation crowd to reach the target intersection, and perform clustering analysis on the evacuation pedestrians on each path respectively;

[0063] The cluster segmentation unit is used to divide the pedestrians on multiple paths into multiple flowing local density clusters respectively through the DBSCAN clustering algorithm;

[0064] And calculate the cluster density, and calculate the density ρ of multiple clusters respectively through the following calculation formula:

[0065]

[0066] Among them, N is the number of pedestrians in a cluster, which can be obtained through an infrared sensor; A is the path area covered by a cluster, which is obtained through a visualization sensor, and when dividing the cluster, ensure that multiple paths from the evacuation location to the target intersection are dynamically covered, that is, within a fixed determination time, any path is completely covered by multiple clusters;

[0067] The calculation unit maps between the path network and the clusters according to the clusters segmented by the cluster segmentation unit, maps with a cluster as a sub-unit on the path network, and calculates the moving speeds of each cluster respectively: by measuring the moving distances of multiple clusters within a unit time respectively, the moving speed of a single cluster is calculated, and according to the moving speeds of multiple clusters and the distance between the location of a specific cluster and the target intersection, calculate the time for multiple clusters to reach the target intersection respectively, and generate a pedestrian-time curve for reaching the target intersection in combination with the number of pedestrians in each cluster;

[0068] As mentioned above, the pedestrians merging into the target intersection also include the real-time pedestrian density data of the current road. The current road here is the other roads centered on the target intersection, excluding the road from the evacuation location to the target intersection, and is the road where pedestrians or vehicles merge.

[0069] The calculation unit fuses the real-time pedestrian density data of the current road with the previously calculated pedestrian-time curve to obtain the arrival intersection pedestrian-time curve of the current intersection, calculates the peak value of the arrival intersection pedestrian-time curve, calibrates the number of pedestrians corresponding to the peak value as the pedestrian peak value, and the time when the peak value is reached as the pedestrian peak appearance time. The calculation unit transmits the predicted pedestrian peak appearance time and the pedestrian peak value to the traffic light control module, and the traffic light control module makes subsequent judgments.

[0070] The calculation unit calculates the number of pedestrians included in each cluster group, and uses statistics to calculate the number of people arriving at the target intersection through multiple paths from the dispersal location, so as to calculate the time when the number of pedestrians from the dispersal location to the target intersection location reaches the peak. ； And by dividing each of the multiple paths from the dispersal location to the target intersection into multiple cluster groups, and calculating the group density and the moving speed of each cluster group, the time of pedestrians on multiple paths can be calculated quickly.

[0071] The path topology data from the dispersal location to the target intersection is obtained by the topology data acquisition unit from the GIS road network data. The collection of the pedestrian flow information collected by the camera or sensor for the pedestrian data comes from different devices and systems. Since both the GIS road network data and the camera collection or sensor collection are set with their own internal clocks, there may be a deviation of milliseconds between these clocks, and there may be a processing delay between the original data collection and the final available data set, which will cause a difference in the time dimension of the two sets of data. The pedestrian data here includes the data about the pedestrians merging into the target intersection obtained by the current road acquisition module and the dispersal data acquisition module, including real-time pedestrian density data, etc.

[0072] Therefore, in order to reduce the difference, in one embodiment, the data prediction module uses the DTW algorithm to find the best time alignment path between the time series of the two sets of data, including the following steps:

[0073] a: Organize the path topology data obtained from the GIS road network and the pedestrian data collected by the camera or sensor into time series respectively. Among them, let the time series of the path topology data be X = {x 1 ,x 2 ,…,x n}, and the time series of the pedestrian data be Y = {y 1 ,y 2 ,…,y m}, and calculate the distance between each data point in the two time series using the Euclidean distance.

[0074] b: Based on the distances calculated in S1, construct an n×m distance matrix D, and the element D(xi , y i ) represents the distance between element x i and element y i ;

[0075] c: Initialize the cumulative cost matrix:

[0076] Create an (n + 1) × (m + 1) cumulative cost matrix C, and initialize its first row and first column to infinity (∞), but C0,0 = 0. The cumulative cost matrix C is used to record the minimum cumulative cost from the starting point to each intermediate point, preparing for subsequent dynamic programming calculations. The purpose of initialization is to ensure that in subsequent calculations, the optimal path is searched starting from the starting point;

[0077] d: Fill the cumulative cost matrix:

[0078] Starting from C1,1, traverse each element Ci,j of the matrix, where i corresponds to sequence A and j corresponds to sequence B, and fill the cumulative cost matrix C row by row and column by column. Use the following formula to calculate the minimum cumulative cost:

[0079] C(i, j) = D(i - 1, j - 1) + min{C(i - 1, j), C(i, j - 1), C(i - 1, j - 1)};

[0080] Among them, C(i, j - 1) represents that the time point in column B is ahead, C(i - 1, j) represents that the time point in row A is ahead, C(i - 1, j - 1) represents that the time in row A and column B is synchronized, and min{C(i - 1, j), C(i, j - 1), C(i - 1, j - 1)} represents the minimum cumulative cost of the previous three paths. i = 1, 2,..., n, j = 1, 2,..., m, so as to continuously recursively calculate the minimum cumulative cost to reach each point, and finally obtain the entire cumulative cost matrix C.

[0081] e: Backtrack to find the optimal alignment matrix:

[0082] Starting from the lower right corner Cn,m of the cumulative cost matrix C, backtrack to the upper left corner C0,0 to find a path with the minimum cumulative cost. This path is the optimal alignment path of the two time series corresponding to the two sets of data. According to the optimal alignment path found in the previous step, align the path topology data time series X and the pedestrian data time series Y to obtain the two sets of aligned data.

[0083] The data prediction module uses the DTW algorithm to perform subsequent calculations after aligning the two sets of data, ensuring the alignment of the two time series in the time dimension and reducing the differences caused by clock deviation and processing delay.

[0084] In one embodiment, a method for a smart city management system includes the following steps:

[0085] S1: Predict the time for the pedestrians leaving the venue to reach the target intersection based on the data of the pedestrians leaving the venue and the road data from the leaving position to the target intersection, and combine the real-time pedestrian density data to predict the time when the peak number of pedestrians reaching the target intersection appears and the peak data. When the time when the peak pedestrian flow data appears arrives, control the current intersection to have a green light for pedestrians;

[0086] S2: Calculate the green light time for pedestrians required for the number of pedestrians corresponding to the peak data to pass through the target intersection according to the peak data and the preset traffic volume of the current intersection, and control the green light for pedestrians to last with the green light time for pedestrians;

[0087] S3: Until the predicted value is less than the peak traffic flow data, control the current intersection to have a green light for vehicles.

[0088] As mentioned above, it is only a preferred embodiment of the present invention and does not impose any form of limitation on the present invention. Although the present invention has been disclosed as above with a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments by using the above-disclosed technical content within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A smart city management system, characterized in that: include; The current road data acquisition module is used to obtain the real-time pedestrian density data on the current sidewalk and the real-time vehicle flow data on the current road; The data acquisition module for dispersing the scene is used to obtain the data of dispersing vehicles and pedestrians, as well as the road data from the dispersing location to the target intersection; The data prediction module is used to predict the time when the pedestrians will arrive at the target intersection based on the pedestrian data and the road data from the location of the pedestrians to the target intersection, and then predict the peak time and peak value of pedestrians at the target intersection in combination with the real-time pedestrian density data; A traffic light control module is used to control the current intersection to be a pedestrian green light when the pedestrian peak time arrives, and is also used to calculate the pedestrian green light time required for the pedestrians with the peak number of pedestrians to pass through the target intersection according to the pedestrian peak time and the preset current intersection traffic volume, and control the lighting time of the pedestrian green light according to the calculated pedestrian green light time; Until the predicted value is less than the pedestrian peak value, the traffic light control module controls the current intersection to be green for vehicles; The traffic light control module calculates the green light time required for the total traffic flow data to pass through the target intersection based on the total traffic flow data and the preset traffic flow data of the target intersection, and uses the vehicle green light time to control the lighting time of the vehicle green light. The total traffic flow data includes real-time traffic flow data and departure traffic flow data.

2. A smart city management system according to claim 1, characterized in that: The data prediction module includes a vehicle flow prediction unit, which is used to predict the peak time of vehicles arriving at the target intersection based on the departure vehicle data and the road data from the exit location to the target intersection in combination with the real-time traffic flow data, and transmit the vehicle peak time to the traffic light control module; If the vehicle peak time is earlier than the end time of the pedestrian green light, the traffic light control module controls the target intersection to be a green light for vehicles when the vehicle peak time arrives.

3. A smart city management system according to claim 2, characterized in that: The data prediction module also includes a topological data acquisition unit, a cluster segmentation unit and a calculation unit: The topological data acquisition unit is used to acquire the coordinates of the dispersing position, the coordinates of the target intersection, and multiple paths from the dispersing position to the target intersection from the GIS road network data; The cluster segmentation unit is used to divide the pedestrians on the plurality of paths into a plurality of flowing local density clusters by using a DBSCAN clustering algorithm; The calculation unit is used to monitor the moving speeds of multiple clusters respectively according to the clusters segmented by the cluster segmentation unit using a target tracking algorithm, and calculate the time it takes for multiple clusters to reach the target intersection in combination with the distances from the real-time positions of the multiple clusters to the target intersection.

4. A smart city management system according to claim 3, characterized in that: The calculation unit calculates the time for the multiple clusters to reach the target intersection according to the moving speeds of the multiple clusters and the distance between the location of the specific cluster and the target intersection, and generates a pedestrian-time curve for reaching the target intersection in combination with the number of pedestrians in each cluster; The calculation unit fuses the real-time pedestrian density data with the pedestrian-time curve to obtain the pedestrian-time curve of the current intersection, and calculates the peak value of the curve, marks the peak value as the pedestrian peak value, and the time when the peak value is reached is the pedestrian peak occurrence time.

5. A smart city management system according to claim 4, characterized in that: The data prediction module uses the DTW algorithm to find the best time alignment path between the time series of two sets of data, including the following steps: a: The path topology data obtained from the GIS road network and the pedestrian data collected by cameras or sensors are organized into time series respectively, and the distance between each data point in the two time series is calculated using the Euclidean distance; b: Construct an n×m distance matrix D; c: Initialize the cumulative cost matrix C, which is (n+1)×(m+1) dimensional: d: Traverse each element C(i, j) of the matrix, where i corresponds to sequence A and j corresponds to sequence B, and fill the cumulative cost matrix C row by row and column by column: e: Backtrack to find the best alignment matrix, and find the best alignment path based on the best alignment matrix, and align the time series of the two sets of data according to the best alignment path.

6. A smart city management system according to claim 5, characterized in that: In step S4, the minimum cumulative cost is calculated using the following formula: C(i,j)=D(i-1,j-1)+min{C(i-1,j),C(i,j-1),C(i-1,j-1)}; Wherein, C(i, j-1) indicates that the time point of column B is ahead, C(i-1, j) indicates that the time point of row A is ahead, C(i-1, j-1) indicates that row A and column B are synchronized in time, min{C(i-1, j), C(i, j-1), C(i-1, j-1)} indicates the minimum cumulative cost of the first three paths, i = 1, 2, ..., n, j = 1, 2, ..., m; The minimum cumulative cost to reach each point is calculated recursively and the cumulative cost matrix C is obtained.

7. A smart city management system according to claim 5, characterized in that: In step S5, the two optimal alignment paths are obtained by starting from the C(n,m) element of the cumulative cost matrix C and tracing back to the starting point C(0,0).

8. A smart city management system according to claim 1, characterized in that: The road data acquisition module includes an infrared induction pedestrian counter, a geomagnetic sensor and a high-definition camera.

9. A method of a smart city management system, using the smart city management system according to claim 1, characterized in that: The following steps are involved: S1: The time when the pedestrians arrive at the target intersection is predicted by using the pedestrian data and the road data from the location of the pedestrians to the target intersection. The peak time and peak data of pedestrians arriving at the target intersection are predicted by combining the real-time pedestrian density data. When the peak data of the pedestrian flow arrives, the current intersection is controlled to have a green light for pedestrians. S2: Calculate the pedestrian green light time required for the number of pedestrians with peak data to pass through the target intersection according to the pedestrian peak value and the preset current intersection traffic volume, and use the pedestrian green light time to control the duration of the pedestrian green light; S3: until the predicted value is less than the pedestrian peak value, the current intersection is controlled to have a green light for vehicles.

Images