A vehicle dispatching system and method for urban management

Abstract

The present invention relates to the technical field of vehicle path scheduling, and in particular to a vehicle scheduling system and method for urban management, the method comprising: obtaining a vehicle position, a destination, and a plurality of road sections to be traveled between the position and the destination; constructing an artificial potential field using the congestion degree of each road section to be traveled; and performing vehicle path planning based on the constructed artificial potential field to complete vehicle path scheduling; that is, the solution of the present invention can improve the accuracy of vehicle path scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle path scheduling, and more specifically, to a vehicle scheduling system and method for urban management. Background Art

[0002] With the rapid increase in the number of private cars and the relatively limited road resources, urban traffic congestion has become increasingly serious. Traffic congestion affects citizens' travel time, increases commuting costs, and leads to frequent traffic accidents. In addition, long-term traffic congestion will also aggravate air pollution. Therefore, alleviating traffic congestion and optimizing traffic flow are the top priorities of urban traffic management.

[0003] At present, with the rapid development of intelligent transportation systems, new opportunities have been brought to solve traffic congestion. Intelligent transportation systems can collect and monitor the real-time road conditions of the road network through sensors, and then provide individual travelers with a path to avoid congestion and travel efficiently through path planning algorithms. Therefore, the research on path planning is very meaningful.

[0004] The intelligent transportation system obtains traffic flow data and uses various algorithms to predict the traffic conditions of the road in the future. Based on this, it provides users with an optimal driving route. At the same time, during the driving process, the intelligent transportation system obtains the traveler’s current location and the current state of the urban road network, predicts the congestion of the roads in the urban road network when entering the next intersection, and dynamically plans the travel route for the traveler in real time.

[0005] For example, the patent document with authorization announcement number CN116311948B, titled Vehicle path planning method based on traffic flow speed prediction and signal light status, discloses predicting the average driving speed of the current time interval based on the average driving speed of multiple historical time intervals for each path to determine the travel time, and determines the action of the target vehicle to travel from the current path to the next path with the help of a trained deep reinforcement learning model.

[0006] In the above scheme, when obtaining the average driving speed, the corresponding traffic flow sequence often contains multiple changing rules, such as changes in traffic flow speed caused by commuting, traffic accidents, and weather changes, and various rules are mixed with each other; using mWDN (mutil Wavelet Decomposition Network) to separate the above changing rules not only makes the calculation complexity high, but also because it is sensitive to noise, it is difficult to determine the impact of the rules corresponding to different influencing factors on the path. Therefore, when conducting subsequent analysis of traffic flow characteristics, there will inevitably be inaccuracies, which will affect vehicle path scheduling. Summary of the invention

[0007] The purpose of the present invention is to propose a vehicle dispatching system and method for urban management, so as to solve the problem that the existing technology has inaccurate analysis of real-time traffic flow data, thereby affecting the vehicle path scheduling; to this end, the present invention provides solutions in the following two aspects.

[0008] In a first aspect, the present invention provides a vehicle dispatching method for urban management, comprising:

[0009] Obtain the vehicle's location, destination, and multiple road sections to be traveled between the location and the destination;

[0010] An artificial potential field is constructed using the congestion level of each road section to be traveled; based on the constructed artificial potential field, the vehicle path is planned to complete the vehicle path scheduling;

[0011] The value of the repulsive field function in the artificial potential field is positively correlated with the congestion degree of each road section to be traveled;

[0012] The congestion degree is: obtaining traffic flow data of all road sections to be traveled; the traffic flow data includes characteristic data of multiple dimensions;

[0013] Obtain historical traffic flow data of all road sections in the area where the vehicle is located, divide all road sections into multiple clusters, and determine the historical congestion level of each cluster;

[0014] Calculate the distance between any road segment to be driven and each cluster. for:

[0015] ; is the cluster center corresponding to the sth cluster Characteristic data; is the ath road segment corresponding to the ath road segment to be traveled feature data, K is the total number of feature data, is the weight of the kth feature data in the sth cluster, and the weight is the normalized value of the mutual information entropy between each feature data in each cluster and the historical congestion degree of the corresponding cluster;

[0016] The historical congestion level corresponding to the cluster with the smallest distance is taken as the congestion level of the road section to be traveled.

[0017] The above scheme utilizes the traffic conditions of the road sections in the historical data to confirm the mutual information entropy of each data in the traffic flow data under different historical congestion levels and the corresponding historical congestion levels, and matches the road section to be traveled with the cluster, and takes the historical congestion level of the cluster closest to the road section to be traveled as the congestion level of the road section; and improves the repulsive field function in the subsequently constructed artificial potential field to realize vehicle path planning, thereby improving the accuracy of path planning.

[0018] Optionally, the repulsive field function is a first repulsive field function, specifically:

[0019] ;

[0020] in, represents the first repulsive field function, represents the repulsion scale factor, Indicates the current position of the vehicle. Indicates the road section to be traveled. is a vector whose magnitude is the distance from the vehicle's current position to the road section to be driven, and whose direction is the direction from the current position to the road section to be driven. is the coefficient of the impact of congestion level on path optimization, Indicates the congestion level of the road section to be traveled.

[0021] The above scheme introduces the congestion level of the road section to be traveled, and can plan a path suitable for the vehicle based on the traffic conditions of the road section to improve the travel efficiency.

[0022] Optionally, the repulsive field function is a second repulsive field function, specifically:

[0023] ;

[0024] ;

[0025] in, represents the second repulsive field function, represents the repulsion scale factor, Indicates the current position of the vehicle. Indicates the road section to be traveled. is a vector whose magnitude is the distance from the vehicle's current position to the road section to be driven, and whose direction is the direction from the current position to the road section to be driven. is the coefficient of the impact of congestion level on path optimization, Indicates the congestion level of the road section to be traveled. Indicates travelers’ dissatisfaction with the road section they travel.

[0026] The above scheme can plan road sections that meet travelers' preferences by introducing travelers' satisfaction with road sections.

[0027] Optionally, the characteristic data of multiple dimensions in the traffic flow data are vehicle volume, vehicle speed, accident rate and number of road lanes.

[0028] Optionally, the method for normalizing the mutual information entropy between each feature data in each cluster and the historical congestion degree of the corresponding cluster includes: a maximum and minimum value normalization method and a Z-Score normalization.

[0029] The above normalization of the mutual information entropy corresponding to different feature data in each cluster can simplify the calculation and improve the calculation efficiency.

[0030] Optionally, the specific process of dividing all road sections to obtain multiple clusters is:

[0031] Based on the historical traffic flow data of all road sections, the K-means clustering algorithm is used to cluster all road sections to obtain multiple clusters.

[0032] The above scheme performs cluster analysis on all road sections in the city where the vehicle is located, and classifies road sections with the same or similar traffic conditions into one category, which is helpful for subsequent correlation analysis of traffic flow data and traffic conditions of road sections under different traffic conditions.

[0033] Optionally, the historical congestion degree is marked according to the average historical travel time of all vehicles in the cluster.

[0034] In a second aspect, a vehicle dispatching system for urban management includes:

[0035] processor;

[0036] The memory stores a computer instruction for vehicle dispatching for city management. When the computer instruction is executed by the processor, the system executes the above-mentioned vehicle dispatching method for city management.

[0037] The beneficial effects of the present invention are:

[0038] The scheme of the present invention can determine the importance of each data for different historical congestion levels by calculating the mutual information entropy between each data and the historical congestion level in the historical traffic flow data of all sections in the urban roads, and can accurately determine the congestion situation of the section to be traveled by the vehicle in combination with the historical congestion level of the historical traffic, further providing accurate data support for the repulsive field function in the subsequently constructed artificial potential field, thereby improving the accuracy of the vehicle's path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0040] Figure 1 A flowchart schematically shows a vehicle dispatching method for city management in this embodiment;

[0041] Figure 2The structural block diagram of a vehicle dispatching system for city management in this embodiment is schematically shown. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0043] The present invention is aimed at path planning of a vehicle during its travel on urban roads.

[0044] A vehicle dispatching method for urban management of the present invention is introduced below by taking a taxi or online-ordered vehicle in motion as an example.

[0045] Specifically, Figure 1 As shown, a vehicle dispatching method for urban management in this embodiment includes the following steps:

[0046] Step S1, obtaining the vehicle's location, destination, and multiple road sections to be traveled between the location and the destination.

[0047] Specifically, the current position of the vehicle is acquired using sensors, wherein the sensors include a laser radar.

[0048] The destination is the final location that the vehicle is going to reach. There may be multiple drivable paths between the current location of the vehicle and the destination, and each path is a combination of multiple road sections.

[0049] It should be noted that the above-mentioned road section can be abstracted into a point or a grid. Taking a point (road section) as an example, the point (road section) has a size and a position, where the size is the distance between the stop lines of two adjacent intersections (intersections refer to signal-controlled intersections), that is, a distance between two intersections; the position of the point (road section) can be any position information of two intersections and between them. In this embodiment, the position information of the midpoint between the two intersections is selected.

[0050] Step S2, constructing an artificial potential field using the pre-acquired congestion levels of each road section to be traveled; and performing vehicle path planning based on the constructed artificial potential field to complete vehicle path scheduling.

[0051] In this embodiment, considering the traffic conditions of each road section in the vehicle's path planning, it is necessary to first calculate the congestion level of the road section to be traveled by the vehicle. The specific process is as follows:

[0052] Step S21, obtaining historical traffic flow data of each section in the urban road network. Specifically, the historical traffic flow data of all sections in the urban road within a set time can be retrieved from the traffic information service system. The historical traffic flow data includes vehicle volume, vehicle speed, accident rate and number of road lanes.

[0053] Specifically, the traffic volume can be collected through the video monitoring system, by identifying the vehicle information in the video frame and tracking and counting it. Specifically, the traffic volume = the number of passing vehicles / time. Another calculation formula is: traffic volume = unit time * vehicle speed / (vehicle distance + vehicle body length); the vehicle speed is collected by a laser speed meter; the accident rate is the ratio of the total number of accidents of all vehicles in the driving section to the total traffic volume or total driving distance, and the number of road lanes is obtained through urban road network information or map navigation.

[0054] The above-mentioned set time can be the rush hour (between 7 am and 9 am), or the rush hour (between 5 pm and 8 pm), or other time periods or the whole day. Specifically, the above-mentioned set time needs to be determined according to the time period in which the vehicle is currently traveling.

[0055] Step S22: Divide all road sections to obtain clusters with different historical traffic congestion levels.

[0056] In this embodiment, a clustering algorithm is used to cluster all road sections to obtain multiple clusters, wherein the clustering algorithm uses a K-means clustering algorithm, a mean clustering algorithm, and the like.

[0057] Taking the K-means clustering algorithm as an example, all road sections are clustered to obtain multiple clusters, specifically:

[0058] A set number of road sections are randomly selected as initial centers, the Euclidean distances from other road sections to each initial center are calculated, and they are assigned to the cluster with the closest initial center; for each cluster, the initial center is updated to obtain the cluster center, and multiple iterations are performed until the location of the cluster center remains unchanged or the preset number of iterations is reached, and multiple clusters are obtained. The value of the set number in this embodiment can be 4; of course, according to actual conditions, it can also be 5, 6 or other values.

[0059] Before calculating the Euclidean distance, it is also necessary to perform a step of normalizing the feature data of each dimension in the historical traffic flow data of each road section, wherein the normalization process may adopt a maximum and minimum value normalization algorithm.

[0060] Since the clustering process of the K-means clustering algorithm is an existing technology, only the important steps are briefly introduced here, and the specific implementation steps are not introduced in detail.

[0061] In this embodiment, after obtaining multiple clusters, it is also necessary to calibrate the traffic congestion conditions of the multiple clusters. Specifically, the historical congestion degree of each cluster is determined based on the historical traffic flow data of each road section in the cluster. Among them, the historical congestion degree in this embodiment can be a congestion level or a specific congestion degree. The congestion level is to classify the traffic conditions, and the congestion degree is to assign different values ​​to different levels of traffic conditions, and the size of the value is in the range of (0,1].

[0062] In one embodiment, the historical congestion degree is calibrated according to the historical travel time average of all historical vehicles in the cluster. For example, when the travel time average exceeds 1h, the traffic situation is "level 4 traffic congestion"; when the travel time average is between 1h-0.5h, the traffic situation is "level 3 traffic congestion"; when the travel time average is between 0.5h-10min, the traffic situation is "level 2 traffic congestion"; when the travel time average is less than 10min, the traffic situation is "level 1 traffic congestion". The higher the traffic congestion level, the more congested the road section.

[0063] In another embodiment, those skilled in the art perform calibration according to actual conditions.

[0064] In the above, the congestion level can also be converted into the corresponding congestion degree, where the higher the traffic congestion level, the greater the congestion degree. For example, the congestion degree of "level 1 traffic congestion" is 0.1, the congestion degree of "level 2 traffic congestion" is 0.3, the congestion degree of "level 3 traffic congestion" is 0.5, and the congestion degree of "level 4 traffic congestion" is 0.8.

[0065] Among them, the congestion level can also be determined based on speed and traffic volume; the speed and traffic volume corresponding to different congestion levels need to be determined based on actual conditions. For example, the speed and traffic volume on expressways are relatively high, while in urban roads, since the distances of each section are not long, the speed and traffic volume are relatively low. Therefore, when judging the congestion situation of urban roads, the speed and traffic volume standards need to be modified.

[0066] Step S23, calculating the distance between any road section to be traveled and each cluster; taking the historical congestion degree corresponding to the cluster with the smallest distance as the congestion degree of the road section to be traveled by the vehicle.

[0067] In this embodiment, it is taken into account that the influence of traffic volume, vehicle speed, accident rate and number of road lanes on the degree of congestion is different. Therefore, the mutual information entropy of each feature data in the historical traffic flow data corresponding to all road sections in each cluster and the corresponding historical congestion degree is calculated to determine the importance of each feature data in the historical traffic flow data, and use it as a weight to match the road section where the vehicle is traveling in real time with the road section with similar historical traffic conditions, so as to determine the congestion degree of the road section to be traveled and improve the accuracy of matching.

[0068] Among them, the mutual information entropy is: ;

[0069] In the formula, is the information entropy of each feature parameter X; is the information entropy of the historical congestion level Y; is the joint distribution probability of each feature parameter X and the historical congestion level Y; It is the mutual information entropy of each feature parameter X and the historical traffic congestion degree Y, where the higher the mutual information entropy value is, the more important the feature parameter is for the calculation of traffic congestion degree.

[0070] Since the calculation of mutual information entropy is an existing technology, it will not be introduced in detail here.

[0071] In one embodiment, after obtaining the mutual information entropy, each mutual information entropy needs to be normalized to obtain a normalized value, and each normalized value is used as the weight of the corresponding feature data, wherein the sum of all weights is 1.

[0072] Specifically, taking the four characteristic data of traffic volume, vehicle speed, accident rate and number of road lanes in the historical traffic flow data as an example, the weight of the kth characteristic data in the sth cluster is for:

[0073] ;

[0074] in, is the first Characteristic data; is the historical congestion level in the sth cluster; is the first The mutual information entropy between a feature and its historical congestion degree.

[0075] As another implementation, the normalization method may be a maximum-minimum value normalization method or a Z-Score normalization method.

[0076] In this embodiment, the distance between each cluster and any road section to be traveled is:

[0077] ; is the cluster center corresponding to the sth cluster Characteristic data; is the ath road segment corresponding to the ath road segment to be traveled feature data, K is the total number of feature data, is the weight of the kth feature data in the sth cluster.

[0078] In the formula, The smaller it is, the more similar the congestion levels of the a-th road segment and the s-th cluster are; The larger the value is, the more important the feature data is for calculating the congestion level in the sth cluster.

[0079] It should be noted that the weights of the characteristic data in different clusters are different. Therefore, when calculating the distance, the weights are different in the calculation of the distance between any road section to be traveled and different clusters.

[0080] In this embodiment, after obtaining the congestion level of the road section to be traveled by the vehicle, the path planning of the vehicle is also performed by constructing an artificial potential field, as follows:

[0081] First, the congestion degree of each road section to be traveled is used to construct an artificial potential field. Specifically, the repulsive field function and the attractive field function are constructed through the artificial potential field method; the value of the repulsive field function is positively correlated with the congestion degree.

[0082] Among them, Artificial Potential Field (APF) is a common path planning and obstacle avoidance method, which achieves path planning and obstacle avoidance by simulating the interaction between objects.

[0083] In one embodiment, the repulsive field function may be a first repulsive field function, which is determined according to the acquired congestion level of the road section to be traveled, and is specifically:

[0084] ;

[0085] in, represents the first repulsive field function, represents the repulsion scale factor, Indicates the current position of the vehicle. Indicates the road section to be traveled. is a vector whose magnitude is the distance from the vehicle's current position to the road section to be driven, and whose direction is the direction from the current position to the road section to be driven. is the coefficient of the impact of congestion level on path optimization, Indicates the congestion level of the road section to be traveled.

[0086] in, The larger the value, the greater the impact of the road section to be traveled on the path optimization.

[0087] As another implementation method, when the vehicle is a taxi or an online car-hailing vehicle, the traveler's satisfaction with the driving section is also introduced to determine the repulsive field function, that is, the repulsive field function of this embodiment can also be a second repulsive field function, specifically:

[0088] ;

[0089] ;

[0090] in, represents the second repulsive field function, represents the repulsion scale factor, Indicates the current position of the vehicle. Indicates the road section to be traveled. is a vector whose magnitude is the distance from the vehicle's current position to the road section to be driven, and whose direction is the direction from the current position to the road section to be driven. is the coefficient of the impact of traffic congestion on path optimization, Indicates the congestion level of the road section to be traveled. Indicates travelers’ dissatisfaction with the road section they travel.

[0091] in, and The larger the value is, the greater the influence of the road section on the path optimization. When the distance between the path and the newly selected path is greater than the influence range of the congestion level and dissatisfaction of the road section, the congestion level and dissatisfaction of the road section have no influence on the newly selected path, so the repulsion field function is 0.

[0092] Among them, the gravitational field function is: ;

[0093] In the formula, represents the gravitational field function, Indicates the current position of the vehicle. Indicates the destination location, Represents a vector whose magnitude is the driving distance from the current location to the destination and whose direction is the distance from the current location to the destination. is the proportional gain coefficient.

[0094] Secondly, after determining the repulsive field function and the gravitational field function, the total potential field function of the artificial potential field is calculated. When the value of the total potential field function is the largest, the optimal road section can be obtained, and then the optimal path for the vehicle can be obtained.

[0095] Among them, since the specific implementation process of the artificial potential field method is an existing technology, this embodiment only introduces the improved repulsive field function, and the others are not repeated.

[0096] When constructing the repulsive field function and the gravitational field function, the congestion level between the current vehicle position and the road ahead is taken into consideration to improve the repulsive field function, thereby obtaining the dynamic repulsive range of the vehicle and improving the accuracy of path planning.

[0097] In one embodiment, the optimal driving route of the taxi under the current traffic conditions can be obtained based on the search results, and the taxi can be dispatched accordingly based on the route information. For example, when the taxi reaches the ath section, it is found that there is level 3 traffic congestion between the bth section and the cth section. The system gives the optimal driving path for the taxi under the current traffic conditions. If the system shows that the traffic between the bth section and the cth section is smooth while driving to the bth section, the system changes back to the original optimal driving path.

[0098] The solution of the present invention can quickly obtain the congestion level of each section in the route that the vehicle is about to travel, and timely avoid it when planning the route, avoiding long waiting times and the inability to reach the destination, thereby improving the vehicle's travel efficiency.

[0099] The present invention also provides a vehicle dispatching system for urban management. Figure 2 As shown, the system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the vehicle dispatching method for urban management according to the present invention is implemented.

[0100] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface, whose configuration and functions are known in the art and thus will not be described in detail here.

[0101] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0102] In the description of this specification, “plurality” means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0103] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A vehicle dispatching method for urban management, characterized in that: include: Obtain the vehicle's location, destination, and multiple road sections to be traveled between the location and the destination; Using the congestion level of each road section to be traveled, an artificial potential field is constructed; Based on the constructed artificial potential field, vehicle path planning is performed to complete vehicle path scheduling; The value of the repulsive field function in the artificial potential field is positively correlated with the congestion degree of each road section to be traveled; The congestion degree is: obtaining traffic flow data of all road sections to be traveled; the traffic flow data includes characteristic data of multiple dimensions; Obtain historical traffic flow data of all road sections in the area where the vehicle is located, divide all road sections into multiple clusters, and determine the historical congestion level of each cluster; Calculate the distance between any road segment to be driven and each cluster. for: ; is the cluster center corresponding to the sth cluster Characteristic data; is the ath road segment corresponding to the ath road segment to be traveled feature data, K is the total number of feature data, is the weight of the kth feature data in the sth cluster, and the weight is the normalized value of the mutual information entropy between each feature data in each cluster and the historical congestion degree of the corresponding cluster; The historical congestion degree corresponding to the cluster with the smallest distance is used as the congestion degree of the road section to be traveled, including calculating the mutual information entropy between each feature data in the historical traffic flow data corresponding to all road sections in each cluster and the corresponding historical congestion degree, so as to determine the importance of each feature data in the historical traffic flow data, and use it as a weight to match the road section where the vehicle is traveling in real time with the road section with similar historical traffic conditions, so as to determine the congestion degree of the road section to be traveled and improve the accuracy of matching; Among them, the mutual information entropy is: ; In the formula, is the information entropy of each feature parameter X; is the information entropy of the historical congestion level Y; is the joint distribution probability of each feature parameter X and the historical congestion level Y; is the mutual information entropy between each feature parameter X and the historical traffic congestion level Y, where the higher the mutual information entropy value, the more important the feature parameter is for the calculation of traffic congestion level; After obtaining the mutual information entropy, each mutual information entropy needs to be normalized to obtain a normalized value, and each normalized value is used as the weight of the corresponding feature data, where the sum of all weights is 1; Specifically, taking the four characteristic data of traffic volume, vehicle speed, accident rate and number of road lanes in the historical traffic flow data as an example, the weight of the kth characteristic data in the sth cluster is for: ； in, is the first Characteristic data; is the historical congestion level in the sth cluster; is the first The mutual information entropy between a feature and its historical congestion degree.

2. A vehicle dispatching method for urban management according to claim 1, characterized in that: The repulsive field function is a first repulsive field function, specifically: ； in, represents the first repulsive field function, represents the repulsion scale factor, Indicates the current position of the vehicle. Indicates the road section to be traveled. is a vector whose magnitude is the distance from the vehicle's current position to the road section to be driven, and whose direction is the direction from the current position to the road section to be driven. is the coefficient of the impact of congestion level on path optimization, Indicates the congestion level of the road section to be traveled.

3. A vehicle dispatching method for urban management according to claim 1, characterized in that: The repulsive field function is a second repulsive field function, specifically: ； ； in, represents the second repulsive field function, represents the repulsion scale factor, Indicates the current position of the vehicle. Indicates the road section to be traveled. is a vector whose magnitude is the distance from the vehicle's current position to the road section to be driven, and whose direction is the direction from the current position to the road section to be driven. is the coefficient of the impact of congestion level on path optimization, Indicates the congestion level of the road section to be traveled. Indicates travelers’ dissatisfaction with the road section they travel.

4. A vehicle dispatching method for urban management according to claim 1, characterized in that: The method for normalizing the mutual information entropy between each feature data in each cluster and the historical congestion degree of the corresponding cluster includes: a maximum and minimum value normalization method and a Z-Score normalization.

5. A vehicle dispatching method for urban management according to claim 1, characterized in that: The specific process of dividing all road sections to obtain multiple clusters is as follows: Based on the historical traffic flow data of all road sections, the K-means clustering algorithm is used to cluster all road sections to obtain multiple clusters.

6. A vehicle dispatching method for urban management according to claim 1, characterized in that: The historical congestion level is marked according to the average historical travel time of all vehicles in the cluster.

7. A vehicle dispatching system for urban management, characterized in that: include: processor; A memory storing a computer instruction for vehicle dispatching for urban management. When the computer instruction is executed by the processor, the system executes a vehicle dispatching method for urban management according to any one of claims 1-6.

Images